SEARCH

SEARCH BY CITATION

Keywords:

  • Balamuthia mandrillaris;
  • granulomatous amoebic encephalitis;
  • metalloproteases;
  • central nervous system;
  • cytotoxicity

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Balamuthia mandrillaris is a recently identified protozoan pathogen that can cause fatal granulomatous encephalitis. However, the pathogenesis and pathophysiology of B. mandrillaris encephalitis remain unclear. Because proteases may play a role in the central nervous system (CNS) pathology, we used spectrophotometric, cytopathic and zymographic assays to assess protease activities of B. mandrillaris. Using two clinical isolates of B. mandrillaris (from human and baboon), we observed that B. mandrillaris exhibits protease activities. Zymographic assays revealed major protease bands of approximate molecular weights in the region of 40–50 kDa on sodium dodecyl sulfate–polyacrylamide gels using gelatin as substrate. The protease bands were inhibited with 1,10-phenanthroline, suggesting metallo-type proteases. The proteolytic activities were observed over a pH range of 5–11 with maximum activity at neutral pH and at 42°C. Balamuthia mandrillaris proteases exhibit properties to degrade extracellular matrix (ECM), which provide structural and functional support to the brain tissue. This is shown by degradation of collagen I and III (major components of collagenous ECM), elastin (elastic fibrils of ECM), plasminogen (involved in proteolytic degradation of ECM), as well as other substrates such as casein and gelatin but not haemoglobin. However, these proteases exhibited a minimal role in B. mandrillaris-mediated host cell death in vitro using human brain microvascular endothelial cells (HBMECs). This was shown using broad-spectrum matrix metalloprotease inhibitors, GM 6001 and GM 1489, which had no effect on B. mandrillaris-mediated HBMEC cytotoxicity. This is the first demonstration that B. mandrillaris exhibits metalloproteases, which may play important role(s) in the ECM degradation and thus in CNS pathology.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The members of free-living amoebae are becoming increasingly important in human health and are known to cause fatal infections involving the central nervous system (CNS). This is due to (1) the increasing numbers of susceptible hosts (immunocompromised patients), (2) the probable wider distribution of amoebae in the environment due to global warming and (3) our increased awareness (Khan, 2003; Marciano-Cabral & Cabral, 2003; Jayasekera et al., 2004; Schuster & Visvesvara, 2004). However, the lack of available clinical diagnostic methods, especially in developing countries, and their successful implementation remains a major challenge. Among many other members of free-living amoebae, Balamuthia mandrillaris was identified in the 1980s (Visvesvara et al., 1990) and was found to be associated with fatal human infections in 1991 (Anzil et al., 1991; Taratuto et al., 1991). We have recently identified the first B. mandrillaris isolate in the UK from a 33-year-old male, who died of granulomatous encephalitis (Jayasekera et al., 2004; White et al., 2004). Balamuthia granulomatous encephalitis (BGE) was characterized by headache, fever, characteristic skin lesions, stiff neck, acute confused state, with multiple cerebral lesions involving extensive right hemispheric oedema and hydrocephalus under computer tomography scan and finally death (Jayasekera et al., 2004; White et al., 2004). More importantly, the patient was negative for syphilis, fungal and mycobacterial infections, and had no history of human immunodeficiency virus (HIV) 1 and 2 or of any immunocompromised status. Although the predisposing factors in contracting BGE are not known, these findings suggest that B. mandrillaris can cause fatal infections in relatively immunocompetent individuals. Because the pathogenesis of BGE is associated with the ability of B. mandrillaris to invade the CNS and penetrate the brain tissue, it is reasonable to predict that this pathogen exhibits protease activities, identification of which is the objective of the present study. Using a range of substrates, we demonstrated that B. mandrillaris exhibits protease activities that can degrade extracellular matrix (ECM) proteins/glycoproteins as well as plasminogen.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Human brain microvascular endothelial cell cultures

Primary brain microvascular endothelial cells from human origin (HBMECs) were used as previously described (Stins et al., 1997; Alsam et al., 2003). The HBMECs were routinely grown on rat-tail collagen-coated dishes in RPMI 1640 containing 10% heat-inactivated fetal bovine serum, 10% Nu-Serum, 2 mM glutamine, 1 mM pyruvate, penicillin (100 U mL−1), streptomycin (100 μg mL−1), nonessential amino acids and vitamins (Invitrogen, Paisley, UK). For cytotoxicity assays, HBMECs were grown in 24-well plates by inoculating 106 cells mL−1 per well. At this cell density, confluent monolayers were formed within 24 h and subsequently used for cytotoxicity assays.

Cultures of Balamuthia mandrillaris

Balamuthia mandrillaris (isolated from baboon brain tissue) was obtained from the American Type Culture Collection (ATCC50209; http://www.atcc.org) as previously described (Jayasekera et al., 2004). In addition, B. mandrillaris was isolated from the first BGE case in the UK and its identity confirmed using PCR and indirect immunofluorescence assays (Jayasekera et al., 2004). Both isolates were routinely grown on HBMECs as food source. Briefly, B. mandrillaris isolates were inoculated in 10 mL RPMI 1640 (105 amoebae per mL) on HBMEC monolayers grown in T-75 tissue culture flasks (Orange Scientific, Braine-l'Alleud, Belgium). Within 48 h, the isolates were grown to approximately 8 × 105 amoebae per mL (6–8 × 106 total number of amoebae in 10 mL;>95% in trophozoite forms) and were used for all subsequent assays.

Cytotoxicity assays using live Balamuthia mandrillaris cultures, their lysates and their conditioned medium

To examine the pathogenic potential of each isolate used in this study, cytotoxicity assays were performed. Briefly, B. mandrillaris isolates (5 × 105 amoebae per well) were incubated with HBMEC monolayers in serum-free medium (RPMI 1640 containing 2 mM glutamine, 1 mM pyruvate and nonessential amino acids) at 37°C in 5% CO2 incubator. To prepare B. mandrillaris lysates, amoebae (5 × 105 amoebae per 500 μL−1 in 1.5 mL centrifuge tubes) were incubated at −80°C for 20 min (or until frozen) and then completely thawed at 37°C, followed by vortexing. This process was repeated four times. Finally, B. mandrillaris lysates were added to HBMEC monolayers. In some experiments, conditioned media were produced by inoculating B. mandrillaris on HBMEC monolayers as described above. Once the B. mandrillaris had completely destroyed the HBMEC monolayers, cell-free medium (conditioned medium) was collected by centrifugation and used for cytotoxicity assays.

Whole B. mandrillaris isolates, their lysates or their conditioned medium were incubated with HBMEC monolayers for up to 24 h. At the end of this incubation period, monolayer disruptions were assessed visually after haematoxylin staining. In addition, supernatants were collected and cytotoxicity was determined by measuring lactate dehydrogenase (LDH) release (cytotoxicity detection kit; Roche Applied Science, Lewes, UK) as previously described (Sissons et al., 2005). Briefly, conditioned media of cocultures of B. mandrillaris and HBMECs were collected and percentage LDH release was determined as follows: % cytotoxicity =(sample value−control value)/(total LDH release−control value) × 100. Control values were obtained from HBMECs incubated alone. Total LDH release was determined from HBMECs treated with 1% Triton-X-100 (weight in volume, w/v). To determine whether cytotoxic factors of B. mandrillaris are proteinaceous in nature, amoebae were heat-inactivated by incubating at 70°C for 45 min followed by cytotoxicity assays as described above. In some experiments, cytotoxicity assays were performed in the presence or absence of broad-spectrum matrix metalloprotease (MMP) inhibitors, i.e. GM 6001 and GM 1489 (Merck Biosciences Ltd, Nottingham, UK).

Spectrophotometric assays

To determine whether B. mandrillaris exhibits proteolytic activities, spectrophotometric assays were performed by modifying previous protocols (Khan et al., 2000). Briefly, B. mandrillaris (106 amoebae) isolates were resuspended in 0.1 mL RPMI 1640 followed by freeze-thawing (4 ×) as described above. Both B. mandrillaris lysates or their conditioned medium were incubated with 1 mL of the substrate solution (containing 6 mg mL−1 gelatin dissolved in 10 mM sodium tetraborate) for 3 min at various temperatures. Reactions were stopped by adding 1 mL of 110 mM trichloracetic acid (TCA). Mixtures were incubated at 37°C for 20 min. Finally, solutions were transferred to cuvettes and their absorbance determined at 275 nm using a Jenway Spectrophotometer. For controls, B. mandrillaris lysates were added following the addition of TCA. Absorbance was then converted into units of protease activity using the equation: units per mL enzyme =(test absorbance−control absorbance × total volume/time of assays × millimolar extinction coefficient × lysate volume). The millimolar extinction coefficient of tyrosine under these conditions is 1.34 and the volume of lysates is 0.1 mL. One unit of enzyme hydrolyses substrate to produce 1 μmol of tyrosine per min.

Zymographic assays

To identify and characterize B. mandrillaris proteases, zymographic assays were performed as previously described (Cao et al., 1998; Khan et al., 2000). Briefly, lysates of B. mandrillaris (5 × 105 amoebae, freeze-thawed four times) or their conditioned medium were mixed (1 : 1) with sample buffer [containing 4% sodium dodecyl sulfate (SDS) but without β-mercaptoethanol] and electrophoresed on SDS-polyacrylamide gels (SDS-PAGE) containing gelatin (1 mg mL−1) (Sigma Laboratories, Poole, Dorset, UK). After electrophoresis, gels were soaked in 2.5% Triton X-100 (w/v) solution for 60 min to remove SDS. Finally, the gels were incubated in a developing buffer (50 mM Tris/HCl, pH 7.5, containing 10 mM CaCl2) at 37°C overnight, rinsed and stained with Coomassie brilliant blue. Areas of gelatin digestion were visualized as nonstaining regions in the gel. In some experiments, samples were pretreated with phenylmethylsulfonylfluoride (PMSF; an irreversible inhibitor of serine proteases; 2 mM final concentration), or 1, 10-phenanthroline (a metalloprotease inhibitor; 10 mM final concentration). As 1,10-phenanthroline is a reversible inhibitor, it was also included in the developing buffer. To determine the optimum temperature for B. mandrillaris protease activities, gels were incubated at 4, 10, 20, 30, 37, 42 and 65°C. In addition, to determine the optimum pH for B. mandrillaris proteases, gels were incubated at pH 3–12. For pH 3 and 4, developing buffers were prepared using citrate-phosphate buffer (70 mM sodium citrate and 60 mM sodium phosphate) containing 10 mM CaCl2; for pH 5 and 6, developing buffers were prepared using N-[2-morpholino]ethanesulfonic acid (MES) containing 10 mM CaCl2; and for pH 7–12, developing buffers were prepared using 50 mM Tris containing 10 mM CaCl2.

In some experiments, we used ECM proteins including collagen I (obtained from bovine achilles tendon; 1 mg mL−1 final concentration) dissolved in buffer A (50 nM TES {N-tris[hydroxymethyl]methyl-2-aminoethane-sulfonic acid}, pH 7.4 plus 0.36 mM CaCl2), collagen III (calf skin; 1 mg mL−1 final concentration) dissolved in buffer B (distilled H2O, pH 3 using acetic acid), elastin (bovine neck ligament; 1 mg mL−1 final concentration) dissolved in buffer C (200 mM Tris, pH 8.8) and plasminogen (rabbit plasma; 1 mg mL−1 final concentration) dissolved in buffer D (20 mM lysine, pH7) as substrate for B. mandrillaris proteases. In addition, we used casein (bovine milk; 1 mg mL−1 final concentration) dissolved in buffer E (60 mM Tris, 90 mM NaCl, pH 7.5) and haemoglobin (bovine erthyrocytes; 1 mg mL−1 final concentration) dissolved in distilled H2O with gentle heat. All substrates were dissolved according to manufacturer's recommendations (Sigma Laboratories).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Whole live Balamuthia mandrillaris (human and baboon) produced severe HBMEC cytotoxicity as compared with their lysates and conditioned medium

To determine the cytotoxic ability of Balamuthia mandrillaris, cytotoxicity assays were performed using HBMECs. We observed that both the human and baboon isolates of B. mandrillaris produce severe HBMEC cytotoxicity within 24 h (Fig. 1a). To determine whether viability of B. mandrillaris is a prerequisite in HBMEC death, cytotoxicity assays were performed using B. mandrillaris lysates. The B. mandrillaris lysates exhibited minimal HBMEC cytotoxicity as compared with whole live B. mandrillaris (Fig. 1a). To determine whether the cytotoxic factors are secreted, B. mandrillaris conditioned medium was obtained and incubated (neat or diluted 1 : 1 with RPMI 1640) with HBMEC monolayers at 37°C for up to 24 h. The results revealed that conditioned medium exhibited minimal HBMEC cytotoxicity as compared with whole live B. mandrillaris (Fig. 1b). Of interest, heat-inactivation of B. mandrillaris lysates abolished their ability to produce HBMEC cytotoxicity, suggesting that cytotoxic factors are proteinaceous in nature (Fig. 1c).

image

Figure 1.  Whole live Balamuthia mandrillaris (human and baboon isolates) produced severe human brain microvascular endothelial cell (HBMEC) cytotoxicity as compared with their lysates and conditioned medium. Both isolates or their lysates (5 × 105 per well) (a) or their conditioned medium (b) were added to HBMEC monolayers prepared in 24-well plates and cultures were incubated in a 5% CO2 incubator at 37°C for 24 h. HBMECs without amoebae were used as controls. After this incubation, cytotoxicity was determined using lactate dehydrogenase assays. Note that whole live B. mandrillaris exhibited severe HBMEC cytotoxicity whereas their lysates or conditioned medium exhibited minimal HBMEC cytotoxicity. Values are mean±standard error of at least three experiments performed in duplicate. (c) To determine whether cytotoxic factors are proteinaceous in nature, B. mandrillaris was heat-inactivated followed by cytotoxicity assays. Note that heat-inactivation abolished B. mandrillaris-mediated HBMEC cytotoxicity.

Download figure to PowerPoint

Whole Balamuthia mandrillaris digested proteins in spectrophotometric assays

To determine the proteolytic ability of B. mandrillaris lysates or their conditioned medium, spectrophotometric assays were performed. The lysates of B. mandrillaris exhibited gelatin degradation (Fig. 2). Similarly, the conditioned medium (B. mandrillaris plus HBMECs) also exhibited protease activities (Fig. 2). Interestingly, conditioned medium from HBMECs alone also exhibited proteolytic activities (Fig. 2). The B. mandrillaris lysates exhibited maximum proteolytic activities whereas their conditioned medium exhibited the least. Overall, these results suggest that B. mandrillaris exhibits protease activities.

image

Figure 2. Balamuthia mandrillaris lysates and their conditioned medium exhibited protease activities in spectrophotometric assays. To determine proteolytic activities in B. mandrillaris or their conditioned medium (CM), spectrophotometric assays were performed. Note that both B. mandrillaris lysates and their conditioned medium exhibited proteolytic activities. Temperature variations revealed that maximum activities are exhibited at 42°C. Error bars represent SEM. One unit of enzyme hydrolyses substrate to produce 1 μmol of tyrosine per min.

Download figure to PowerPoint

Balamuthia mandrillaris (human and baboon isolates) lysates exhibit several metalloproteases

From spectrophotometric assays, it was confirmed that B. mandrillaris lysates and their conditioned medium exhibit proteolytic activities. To characterize proteases of B. mandrillaris, and to exclude HBMEC proteases, zymography assays were performed using gelatin as substrates. We observed that B. mandrillaris lysates exhibited several protease bands of approximate molecular weights 40–50 kDa (Fig. 3a). Both human and baboon isolates of B. mandrillaris exhibited gelatinase activities. Pretreatment of B. mandrillaris lysates with 1,10-phenanthroline abolished protease activities (Fig. 3b), whereas PMSF had no effect, confirming that B. mandrillaris proteases are metalloproteases (Fig. 3a,b). The conditioned medium exhibited major metalloprotease activities of approximate molecular weight 73 kDa and minimal metalloprotease activities in the region of 40–50 kDa (Fig. 3a). However, the 73 kDa metalloprotease was also observed in conditioned medium from HBMECs alone. Overall, we observed that B. mandrillaris exhibits metalloproteases of approximate molecular weights 40–50 kDa. In addition, we tested the ability of B. mandrillaris proteases to degrade casein and haemoglobin. The results revealed that B. mandrillaris destroyed casein (Fig. 4a) but had no effects on haemoglobin (Fig. 4b).

image

Figure 3. Balamuthia mandrillaris lysates (human and baboon isolates) exhibit several metalloproteases. Both isolates (105 amoebae per lane) and their conditioned medium (15 μL per lane) were electrophoresed on sodium dodecyl sulfate (SDS) polyacrylamide gels (SDS-PAGE) containing gelatin as substrate. Following this, SDS was removed by washing in 2.5% Triton-X-100 and the gel was incubated overnight in developing buffer (50 mM Tris, pH 7.5) containing 10 mM CaCl2. Note that whole B. mandrillaris lysates exhibited several protease bands whereas their conditioned media showed minimal proteolytic activity. Moreover, pretreatment of B. mandrillaris lysates or their conditioned media with phenylmethylsulfonylfluoride (PMSF; a serine protease inhibitor) had no effect on B. mandrillaris proteases (a). By contrast, pretreatment of B. mandrillaris lysates or their conditioned media with 1,10-phenanthroline (a metalloprotease inhibitor), followed by gel incubation in the presence of metalloprotease inhibitor, abolished proteolytic activities of B. mandrillaris (b). Results are representative of three independent experiments.

Download figure to PowerPoint

image

Figure 4. Balamuthia mandrillaris (human and baboon isolates) proteases degrade casein but had no effect on haemoglobin. The ability of B. mandrillaris to degrade casein and haemoglobin was determined using zymographic assays. Note that B. mandrillaris destroys casein, as observed by the nonstaining regions of the gel (a) but had no effects on haemoglobin (b). Results are representative of three independent experiments.

Download figure to PowerPoint

The effects of pH and temperature on the proteolytic activities of Balamuthia mandrillaris metalloproteases

To determine the physiological properties of B. mandrillaris proteases, zymographic assays were performed using gelatin as substrates and gels incubated at various temperatures and pH. The results revealed that B. mandrillaris proteases exhibit metalloprotease activities between pH 5 and 11 (Fig. 5a). However, the optimal metalloprotease activities were observed at neutral pH (Fig. 5a). The influence of temperature on metalloprotease activities were tested at 4–65°C. Optimal activity was observed at 37–42°C (Fig. 5b), indicating their physiological relevance. The metalloproteases lost approximately 50% of optimal activity during incubation at 65°C (Fig. 5b).

image

Figure 5. Balamuthia mandrillaris (human and baboon isolates) proteases exhibit optimum activities at 37°C and at neutral pH. To determine the effects of pH and temperature, zymographic assays were performed and gels were incubated at various pHs (a) and temperatures (b). Optimum proteolytic activities were observed at neutral pH and 37–42°C. Results are representative of three independent experiments.

Download figure to PowerPoint

Balamuthia mandrillaris lysates (human and baboon isolates) exhibit collagenase and elastinolytic activities

The fibrillary collagens and elastin are major components of the ECM. To determine the ability of B. mandrillaris to exhibit collagenase activities, zymographic assays were performed using collagen I and III as substrates. We observed that B. mandrillaris lysates (human and baboon isolates) degraded both collagen I and III substrates (Fig. 6a), confirming that B. mandrillaris exhibits collagenase activities.

image

Figure 6. Balamuthia mandrillaris (human and baboon isolates) proteases degraded extracellular matrix proteins, i.e. collagen I, collagen III and elastin. To determine the collagenolytic and elastinolytic activities of B. mandrillaris, zymographic assays were performed. Note that B. mandrillaris destroyed collagen I (a), collagen III (b) and elastin (c), as observed by the nonstaining regions of the gel. Results are representative of three independent experiments.

Download figure to PowerPoint

Next we determined the ability of B. mandrillaris lysates to degrade elastic fibres. Using elastin (a major constituent of elastic fibres) as substrate in the zymographic assays, we observed that B. mandrillaris exhibits elastinolytic activities (Fig. 6b).

Balamuthia mandrillaris lysates (human and baboon isolates) degrade plasminogen

Other modes of ECM degradation are tissue-type and urokinase-type plasminogen activators. The former pathway is mostly involved in fibrinolysis whereas the latter has been shown to be an important modulator in the pathophysiology of neuronal damage (Busch et al., 1997). The activated urokinase converts proenzyme plasminogen into plasmin, a serine protease involved in ECM destruction via the degradation of fibrin. Here, we determined the ability of B. mandrillaris lysates to degrade plasminogen. The zymographic assays revealed that B. mandrillaris degraded plasminogen at neutral pH (Fig. 7).

image

Figure 7.  Representative effects of Balamuthia mandrillaris (human and baboon isolates) on plasminigen degradation. The ability of B. mandrillaris proteases to degrade plasminogen (a serine protease involved in extracellular matrix destruction) was determined. Plasminogen was used as substrate in zymography assays. We observed that B. mandrillaris proteases degrade plasminogen, as observed by the nonstaining regions of the gel.

Download figure to PowerPoint

Matrix metalloprotease inhibitor, GM 6001 and GM 1489, had no effect on Balamuthia mandrillaris-mediated HBMEC cytotoxicity

From the above, it is clear that B. mandrillaris exhibits proteolytic activities. To determine the cytotoxic ability of proteases, cytotoxicity assays were performed using whole live B. mandrillaris, its lysates and conditioned medium in the presence or absence of matrix metalloprotease (MMP) inhibitors. We observed that both broad-spectrum MMP inhibitors had no effects on B. mandrillaris-mediated HBMEC cytotoxicity (Table 1). Overall, these studies have shown that B. mandrillaris exhibits protease activities, which may play important role(s) in ECM tissue invasion and/or degradation but do not produce HBMEC cytotoxicity.

Table 1.   Metalloprotease inhibitors exhibited no effects on Balamuthiamandrillaris-mediated human brain microvascular endothelial cell cytotoxicity
 HBMEC cytotoxicity (%)
  1. Results are given as mean±SEM.

  2. CM, conditioned medium; HBMEC, human brain microvascular endothelial cell.

Balamuthia mandrillaris72.9±3.4
B. mandrillaris+GM 6001 (5 μg)78.4±4.3
B. mandrillaris+GM 6001 (10 μg)76±7.6
B. mandrillaris+GM 1489 (5 μg)72.2±3.1
B. mandrillaris+GM 1489 (10 μg)70.4±4.1
CM12.3±1.4
CM+GM6001 (5 μg)13.4±1.4
CM+GM1489 (5 μg)11±1.5
B. mandrillaris lysates16.8±2.2
B. mandrillaris lysates+GM6001 (5 μg)15±2.4
B. mandrillaris lysates+GM1489 (5 μg)17.6±2.5

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Balamuthia mandrillaris is a recently identified protozoan pathogen that can cause serious human infections. Infections of the skin or lungs can last for months but the involvement of the CNS almost always lead to death within a few days. This is due to the inability of the host immune system to control infections in the CNS. The pathophysiological complications involving the CNS most probably include invasion of the connective tissue, induction of extensive proinflammatory responses and neuronal damage leading to brain dysfunction. In healthy brain tissue, ECM comprises a major percentage of the normal brain volume (Gladson, 1999), and forms the basal lamina around the blood vessels. The ECM is constantly remodelled and provides the critical structural and functional support to the neuronal tissue. These properties of ECM are tightly regulated by a family of mostly Ca2+-dependent Zn2+-containing endopeptidases (MMPs) (Lukes et al., 1999; Rosenberg, 2002). The ECM plays important roles under normal physiological conditions in development and in maintaining homeostasis in neuronal tissue. However, in neurological disease states, ECM may undergo substantial modifications, resulting in neuroinflammatory responses. The excessive ECM degradation affects neurovascular structural/functional properties that are highly destructive to the CNS functions. ECM is composed of both collagen types, and noncollagenous glycoproteins and proteoglycans (Lukes et al., 1999; Rosenberg, 2002). Thus, the ability of the amoebae to degrade ECM may aid in their invasion of and growth in the brain tissue.

Collagen is the primary component of the ECM and is difficult to degrade owing to its helical structure. The MMPs are the major proteases capable of collagen degradation at neutral pH (Visse & Nagase, 2003). For example, cysteine proteases have been shown to destruct collagen but only at pH 5 (Hou et al., 2001). In our study, we observed that B. mandrillaris exhibits protease properties similar to MMPs and is able to cleave type I and III collagen at neutral pH. These findings suggest that B. mandrillaris proteases may play a role in ECM destruction. This was further confirmed using a metalloprotease inhibitor, i.e. 1,10-phenanthroline, which completely abolished the protease activities. However, a serine protease inhibitor, i.e. PMSF, had no effects on collagen degradation. To our knowledge this is the first report demonstrating proteolytic activities of B. mandrillaris. Moreover, B. mandrillaris metalloproteases exhibited elastinolytic activities. This was demonstrated by the degradation of elastin as substrate in SDS-PAGE. Previous studies have shown that elastase destroys ECM, which increases blood–brain barrier permeability, thereby resulting in brain injury (Janoff, 1985; Rudolphus et al., 1992; Yasui et al., 1995). For example, injection of elastase into the cerebrospinal fluid (CSF) opened the blood–brain barrier in newborn piglets (Temesvari et al., 1995).

In addition to the aforementioned, the urokinase plasminogen activator system plays an important role in various neuronal diseases involving CNS inflammation and/or pathology. For example, in bacterial meningitis, the uPA (urokinase-type plasminogen activator) or tPA (tissue-type plasminogen activator) are known to convert plasminogen (abundant in brain, Soreq & Miskin, 1981) into plasmin, which destroys ECM directly by degrading fibrin or by activating MMPs. Our results demonstrated that B. mandrillaris (human and baboon isolate) directly degraded proenzyme plasminogen, suggesting that pathogenesis of BGE may involve the uPA/tPA system. Again, it has been shown that CSF injection of plasmin results in increased capillary permeability in rats (Armao et al., 1997). Further studies are in progress to determine the activation of plasminogen in response to B. mandrillaris metalloproteases and their targets in neuronal tissue.

The effects of pHs and temperatures on B. mandrillaris metalloproteases were also determined. Thermostability of B. mandrillaris metalloproteases exhibited optimum activity at 37–42°C. The pH optimum of B. mandrillaris metalloproteases indicated that they are neutral/alkaline metalloproteases, highlighting their physiological relevance. One of the interesting findings observed in this study was the levels of secreted proteases in the conditioned medium (Fig. 2). The conditioned medium from HBMECs alone exhibited greater protease activity than the conditioned medium from B. mandrillaris. There are several explanations for these findings: (1) B. mandrillaris proteases are associated with the whole cells and are not secreted in the conditioned medium; (2) once released, B. mandrillaris proteases are inactivated by either host or pathogen factors; (3) proteases are only secreted upon contact with the host cells; or (4) they are membrane-bound. Future studies are in progress to address these issues. Overall, these studies suggest that B. mandrillaris exhibits metalloprotease activities.

The metalloproteolytic enzymes have been identified as virulence factors in various opportunistic pathogens, including Vibrio vulnificus (Miyoshi et al., 1987), Pseudomonas aeruginosa (Wretlind & Pavlovskis, 1983), Aspergillus fumigatus (Markaryan et al., 1994) and Acanthamoeba castellanii (Alsam et al., 2005), and their role as therapeutic targets is under intense investigations. In summary, our findings suggest that B. mandrillaris exhibits metalloprotease activities with the ability to degrade various substrates, and thus may play role(s) in ECM degradation. A complete understanding of the precise mechanisms associated with the role of proteases in the pathogenesis of BGE will help to identify potential targets in the development of therapeutic interventions.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We are grateful to Drs Ed Jarroll (Northeastern University, Boston, MA) and David Warhurst (London School of Hygiene and Tropical Medicine, London, UK) for their advice and support. This work was supported by grants from the Faculty Research Grant, Birkbeck College, University of London.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Alsam S, Kim KS, Stins M, Rivas AO, Sissons J & Khan NA (2003) Acanthamoeba interactions with human brain microvascular endothelial cells. Microb Pathogen 35: 235241.
  • Alsam S, Stins M, Kim KS, Sissons J & Khan NA (2005) Extracellular proteases of pathogenic Acanthamoeba contribute to increased permeability of the blood–brain barrier. J Infect 51: 150156.
  • Anzil AP, Rao C, Wrzolek MA, Visvesvara GS, Sher JH & Kozlowski PB (1991) Amebic meningoencephalitis in a patient with AIDS caused by a newly recognized opportunistic pathogen. Leptomyxid ameba. Arch Pathol Lab Med 115: 2125.
  • Armao D, Kornfeld M, Estrada EY, Grossetete M & Rosenberg GA (1997) Neutral proteases and disruption of the blood–brain barrier in rat. Brain Res 5: 259264.
  • Busch E, Kruger K, Fritze K, Allegrini PR, Hoehn-Berlage M & Hossmann KA (1997) Blood–brain barrier disturbances after rt-PA treatment of thromboembolic stroke in the rat. Acta Neurochir 70 (suppl), 206208.
  • Cao Z, Jefferson. DM & Panjwani N (1998) Role of carbohydrate-mediated adherence in cytopathogenic mechanisms of Acanthamoeba. J Biol Chem 273: 1583815845.
  • Gladson CL (1999) The extracellular matrix of gliomas: modulation of cell function. J Neuropathol Exp Neurol 58: 10291040.
  • Hou WS, Li Z, Gordon RE, Chan K, Klein MJ, Levy R, Keysser M, Keyszer G & Bromme D (2001) Cathepsin k is a critical protease in synovial fibroblast-mediated collagen degradation. Am J Pathol 1159: 21672177.
  • Janoff A (1985) Elastase in tissue injury. Annu Rev Med 36: 207216.
  • Jayasekera S, Sissons J, Tucker J, Rogers C, Nolder D, Warhurst D, Alsam S, White JM, Higgins EM & Khan NA (2004) Post-mortem culture of Balamuthia mandrillaris from the brain and cerebrospinal fluid of a case of granulomatous amoebic meningoencephalitis, using human brain microvascular endothelial cells. J Med Microbiol 53: 10071012.
  • Khan NA (2003) Pathogenesis of Acanthamoeba infections. Microb Pathogen 34: 277285.
  • Khan NA, Jarroll EL, Panjwani N, Cao Z & Paget TA (2000) Proteases: a marker for differentiation of pathogenic and non-pathogenic Acanthamoeba. J Clin Microbiol 38: 28582861.
  • Lukes A, Mun-Bryce S, Lukes M & Rosenberg GA (1999) Extracellular matrix degradation by metalloproteinases and central nervous system diseases. Mol Neurobiol 19: 267284.
  • Marciano-Cabral F & Cabral G (2003) Acanthamoeba spp. as agents of disease in humans. Clin Microbiol Rev 16: 273307.
  • Markaryan A, Morozova I, Yu H & Kolattukudy PE (1994) Purification and characterization of an elastinolytic metalloprotease from Aspergillus fumigatus and immunoelectron microscopic evidence of secretion of this enzyme by the fungus invading the murine lung. Infect Immun 62: 21492157.
  • Miyoshi N, Shimizu C, Miyoshi S & Shinoda S (1987) Purification and characterization of Vibrio vulnificus protease. Microbiol Immunol 31: 1325.
  • Rosenberg GA (2002) Matrix metalloproteinases in neuroinflammation. Glia 39: 279291.
  • Rudolphus A, Stolk J, Van Twisk C, Van Noorden CJ, Dijkman JH & Kramps JA (1992) Detection of extracellular neutrophil elastase in hamster lungs after intratracheal instillation of E. coli lipopolysaccharide using a fluorogenic, elastase-specific synthetic substrate. Am J Pathol 141: 153160.
  • Schuster FL & Visvesvara GS (2004) Free-living amoebae as opportunistic and non-opportunistic pathogens of humans and animals. Int J Parasitol 34: 10011027.
  • Sissons J, Kim KS, Stins M, Jayasekera S, Alsam S & Khan NA (2005) Acanthamoeba induces host cell death via a phosphatidylinositol 3-kinase (PI3K)-dependent mechanism. Infect Immun 73: 27042708.
  • Soreq H & Miskin R (1981) Plasminogen activator in the rodent brain. Brain Res 216: 361374.
  • Stins MF, Gilles F & Kim KS (1997) Selective expression of adhesion molecules on human brain microvascular endothelial cells. J Neuroimmunol 76: 8190.
  • Taratuto AL, Monges J, Acefe JC, Meli F, Parades A & Martinez AJ (1991) Leptomyxid amoeba encephalitis: report of the first case in Argentina. Trans R Soc Trop Med Hyg 85: 77.
  • Temesvari P, Abraham CS, Gellen J Jr, Speer CP, Kovacs J & Megyeri P (1995) Elastase given intracisternally opens blood–brain barrier in newborn piglets. Biol Neonate 67: 5963.
  • Visse R & Nagase H (2003) Matrix metalloproteinases and tissue inhibitors of metalloproteinases: structure, function, and biochemistry. Circ Res 92: 827839.
  • Visvesvara GS, Martinez AJ, Schuster FL, Leitch GJ, Wallace SV, Sawyer TK & Anderson M (1990) Leptomyxid ameba, a new agent of amebic meningoencephalitis in humans and animals. J Clin Microbiol 28: 27502756.
  • White JM, Barker RD, Salisbury JR, Fife AJ, Lucas SB, Warhurst DC & Higgins EM (2004) Granulomatous amoebic encephalitis. Lancet 364: 220.
  • Wretlind B & Pavlovskis OR (1983) Pseudomonas aeruginosa elastase and its role in pseudomonas infections. Rev Infect Dis 5 (suppl): S998S1004.
  • Yasui S, Nagai A, Aoshiba K, Ozawa Y, Kakuta Y & Konno K (1995) A specific neutrophil elastase inhibitor (ONO-5046.Na) attenuates LPS-induced acute lung inflammation in the hamster. Eur Respir J 8: 12931299.