| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
,
* Promics Ltd., The University of Queensland, Brisbane, Australia;
Queensland Brain Institute, The University of Queensland, Brisbane, Australia;
School of Biomedical Sciences, The University of Queensland, Brisbane, Australia; and
Neurology Department, University of Sheffield, Medical School, Sheffield, UK
1 Correspondence: Promics Ltd., The University of Queensland, Brisbane, QLD 4072, Australia. E-mail: trent.woodruff{at}promics.com
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
with ibuprofen or infliximab had no effect in this model. The C5a antagonists did not affect 3-NP-induced cell death when added directly to rat striatal neuronal cultures, indicating a secondary mechanism of action in vivo. Our findings demonstrate for the first time that complement activation in the brain, particularly C5a, is a key event in the pathogenesis of this disease model, and suggest a future role for inhibitors of C5a in the treatment of neurodegenerative diseases.—Woodruff, T. M., Crane, J. W., Proctor, L. M., Buller, K. M., Shek, A. B., de Vos, K., Pollitt, S., Williams, H. M., Shiels, I. A., Monk, P. N., Taylor, S. M. Therapeutic activity of C5a receptor antagonists in a rat model of neurodegeneration.
Key Words: complement system • inflammation • 3-nitropropionic acid
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The role of the complement system and C5a in the periphery has been well characterized over the last few decades (2)
. In addition, cells of the brain are able to synthesize all components and receptors of the complement system based on seminal work in recent years by the individual groups of Scott Barnum, Pat McGeer, and Philippe Gasque/Paul Morgan (3
4
5)
. However, the role of the complement system in general and particularly for C5a in the normal and diseased central nervous system (CNS) is not well understood, and debate remains as to whether it plays a protective or deleterious role in specific disease states (5
6
7)
.
Complement components are up-regulated in numerous neurodegenerative disease states, including Alzheimer’s disease and Huntington’s disease (HD), suggesting a contribution to disease pathology (8
, 9)
. While the role of C5a in neurodegenerative diseases has been examined less, C5aRs appear to be down-regulated in the hippocampus of Alzheimer’s patients, up-regulated in the caudate of Huntington’s patients (9
, 10)
, and up-regulated in animal models of neurological trauma (11
12
13)
. However, determination of the specific role for C5a in neurodegenerative diseases has been hampered by a lack of specific inhibitors of C5a function (6)
.
In recent years there has been considerable interest in the development of complement inhibitors to target certain pathologies, including those of the CNS (7)
. Our group has developed potent and selective small molecule, cyclic peptide C5aR antagonists PMX53 (AcF-[OP(D-Cha)WR]) and PMX205 (hydrocinnamate-[OP(D-Cha)WR]) (14
, 15)
. These antagonists have been shown by us and others to have therapeutic effects in numerous inflammatory disease models in mice and rats, including rheumatoid arthritis; ischemia-reperfusion injuries of the gut, kidney, liver, and limb; inflammatory bowel disease; sepsis; ruptured abdominal aortic aneurysm; liver fibrosis; antiphospholipid antibody (Ab) -induced fetal injury; and traumatic brain cryoinjury (16
17
18
19
20
21
22
23)
. One major advantage of these compounds is their oral activity; PMX53 has been shown to be safe and well tolerated in Phase I clinical trials after either oral or topical administration, making these compounds promising tools for research as well as for further clinical development (15)
.
In this study we hypothesized that neuroinflammation, through complement activation and subsequent C5a generation, plays a role in the pathology of the 3-nitropropionic acid (3-NP) model of striatal degeneration. When this inhibitor of the mitochondrial citric acid cycle is administered via osmotic minipumps to Lewis rats, selective bilateral striatal lesions are formed, mimicking the pathophysiological hallmarks of HD (24)
. 3-NP intoxicated animals also display many of the motor abnormalities seen in patients with HD such as dystonia, bradykinesia, and spontaneous choreiform (25
, 26)
. Consequently, this model has been used in recent years to delineate the mechanisms of pathology in HD and to explore the neuroprotective potential of various therapeutic compounds (27
, 28)
. However, the effect of inhibiting complement or other inflammatory pathways in this model has not yet been investigated. The present study examined the effect of C5aR antagonists (PMX53 and PMX205), a nonsteroidal anti-inflammatory drug (ibuprofen), and a TNF- inhibitor (infliximab) on disease progression in the 3-NP model. The results of this study demonstrate that complement activation—in particular, C5a receptor activation—is involved in the pathology seen in this model and that inhibitors of C5a function could be beneficial in treating neurodegenerative diseases.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The cyclic peptide C5aR antagonists AcF-[OP(D-Cha)WR] (AcPhe[Orn-Pro-D-cyclohexylalanine-Trp-Arg]) and hydrocinnamate-[OP(D-Cha)WR] (hydrocinnamate-[Orn-Pro-D-cyclohexylalanine-Trp-Arg]) were synthesized as described (15)
. These antagonists were dissolved in distilled water (10 mg/ml) for oral dosing, via gavage, and for in vitro studies. Ibuprofen (sodium salt) was dissolved in distilled water (300 mg/L) for oral dosing via drinking water (16)
and for in vitro studies. Infliximab (Remicade; Schering-Plough, Australia) was dissolved in saline (5 mg/ml) per the manufacturer’s instructions for intravenous (i.v.) dosing via the femoral vein (18)
and for in vitro studies.
3-NP model
All animal experiments were approved by The University of Queensland Animal Ethics Committee. Male Lewis rats were purchased from the Animal Resources Center (Australia). The toxin 3-NP was delivered using osmotic minipumps (Alzet, 2ML1; BioScientific, Australia) following a protocol similar to that reported by Ouray and colleagues (24)
. Briefly, a solution of 3-NP (95 mg/ml) was made in 0.1M PBS, then adjusted to pH 7.4 with 5N sodium hydroxide and sterile filtered. On the day of experimentation, this solution was diluted (using 0.1 M sterile PBS) to each rat’s specific body weight, so that each rat received exactly 42 mg/kg 3-NP/day via the osmotic minipumps. Experiments were conducted in groups of 15 rats at a time randomly assigned to treatment groups prior to surgery. Rats were then anesthetized (8 mg/kg xylazine, 20 mg/kg ketamine, 5 mg/kg tiletamine, 5 mg/kg zolazapam; Ilium, Troy Laboratories, Australia; Virbac, Australia) and osmotic pumps were inserted subcutaneously into the back of the animals using sterile surgical techniques. Rats were then returned to individual cages and allowed to recover. On occasion, rats did not recover from anesthetization and surgery, and needed to be removed from the study. The model progressed over 7 days with daily monitoring (measured at the same time point each day) of body weights and food intake (solid food only). Food was freely supplied within the cages due to the rats’ akinesia as a result of 3-NP intoxication. Neurological scores were also performed daily using an established score system that measures gait disturbances, balance, hind-limb dystonia, and recumbency (maximum score of 8; ref 24
). On day 7, rats were euthanized with sodium pentobarbitone (80 mg/kg, intraperitoneal, Nembutal; Rhone Merieux, Australia) and immediately perfused transcardially with 2% sodium nitrite solution (in 0.1 M phosphate buffer, pH 7.4), followed by 500 ml of 4% formaldehyde (in 0.1 M phosphate buffer, pH 7.4). The brains were then removed and processed (see below).
Treatment groups
Large group sizes were used due to the variation in pathology that may be seen between rats, and were as follows. 1) Untreated disease control animals (n=28) were orally dosed with distilled water alone on a daily basis beginning 2 days before the insertion of pumps (day –2); 2) PMX53 (AcF-[OP(D-Cha)WR]) -treated rats (n=22) were orally dosed daily (10 mg/kg, oral) from day –2; 3) PMX205 (hydrocinnamate-[OP(D-Cha)WR]) -treated rats were orally dosed daily (10 mg/kg, oral) either from day –2 (n=19); or 4) 2 days after the insertion of the pumps (post-treatment, n=12); 5) infliximab-treated rats (n=10) were dosed once i.v. on day –2 (5 mg/kg, i.v.); 6) ibuprofen-treated rats (n=12) were orally dosed daily (30 mg/kg, oral) from day –2; and 7) sham-operated animals (n=6) had osmotic mini pumps inserted that contained PBS alone.
A separate group of untreated 3-NP-administered rats (n=14) was also euthanized and perfused at daily intervals over the 7 day period to determine the pathological course of the striatal pathology.
Histological analysis
Fixed brains were paraffin embedded and 5 µm sections were obtained at regular 200 µm intervals throughout the entire striatal region. Sections were then stained with Nissl (0.1% cresyl violet) and with hematoxylin and eosin stains. Nissl sections were examined via light microscopy and photos were taken through the section showing the largest lesion. The average area of the largest lesion size of each animal was then determined by two independent, blinded observers using computerized software (Image J, v. 1.32j; NIH, USA) and expressed as number of pixels within the lesion. Hematoxylin and eosin sections were examined by light microscopy and scored from 0 to 3 by an independent trained observer in a blinded fashion based on the degree of neutrophil infiltration, perivascular edema, hemorrhage, and the extent of striatal degeneration. The total of these scores were then recorded for each individual animal (maximum score of 12).
Apoptosis and neutrophil determination
Striatal sections were deparaffinized, then examined for the degree of apoptosis using a standard TUNEL apoptosis kit (ApopTag; Chemicon, Australia) according to the manufacturer’s instructions. The number of apoptotic cells in the striatal sections of different treatment groups (n=3–6/group) was then counted in a blinded fashion. For neutrophil identification, striatal sections were deparaffinized, then stained for naphthol esterase, a neutrophil-specific marker, using a standard kit (naphthol AS-D chloroacetate esterase).
Immunohistochemistry
After perfusion, some brains were postfixed for 2 h and cryoprotected overnight in 10% sucrose (in 0.1 M PBS, pH 7.4) at 4°C. Serial, coronal forebrain (40 µm) sections were cut using a freezing microtome. Forebrain sections were then exposed to an immunoperoxidase technique to visualize C5aR, C5L2, C3, C9, and GFAP. Briefly, sections were incubated overnight in either mouse anti-C5aR (1:1000), rabbit anti-C5L2 (1:1000), rabbit anti-C3 (1:1000), rabbit anti-C9 (1:1000), and rabbit anti-GFAP (1:1000). This was followed by incubation (2 h) in either biotinylated donkey anti-mouse (1:300; Jackson ImmunoResearch, West Grove, PA, USA) or biotinylated donkey anti-rabbit (1:300; Jackson ImmunoResearch, USA) as appropriate. Sections were incubated for another 2 h in a solution of avidin-biotin-horseradish peroxidase complex (Vector Elite Kit, USA) before being exposed to a nickel 3, 3-diaminobenzidine solution to visualize the immunolabeling. After an overnight wash in PBS, forebrain sections were incubated for 2 h in a biotinylated goat anti-horse (1:300, Jackson ImmunoResearch), which recognizes donkey antibodies. After another 2 h incubation in avidin-biotin-horseradish peroxidase complex, all sections were again exposed to a nickel 3, 3-diaminobenzidine solution. In this way the intensity of the labeling was greatly increased. All forebrain and brainstem sections were then mounted on chrome-alum subbed slides, dehydrated in alcohol, cleared in xylene, and coverslipped. All immunohistochemistry assays included negative controls where the primary Ab was omitted. Upon completion of staining, all slides were coded so that the examiner was blind to the treatment group.
Brain pharmacokinetics of C5aR antagonists
Due to the brains from 3-NP-administered rats being used for histological analysis, in order to determine whether the C5aR antagonists can cross the blood-brain barrier, a separate group of rats (male Wistars, 300–350 g) were anesthetized (8 mg/kg xylazine, 10 mg/kg tiletamine, 10 mg/kg zolazapam) and a 3 mg/kg solution of PMX53 or PMX205 infused i.v. via the femoral vein. Twenty minutes after infusion, a blood sample was collected, and plasma was extracted and stored for later pharmacokinetic analysis. Rats were then immediately perfused with 100 ml of saline to remove circulating blood from the brain. Rat’s brains were then removed and homogenized with 2 ml of distilled water, centrifugated, then the supernatant was stored for pharmacokinetic analysis. Drug levels in the plasma were determined via LC-MS using standard methods as described (15)
, and results were expressed as ng/ml. Drug levels in the brain homogenate were also determined via standard LC-MS methods, with results corrected for extraction efficiency and expressed as ng/gram brain tissue.
Striatal neuronal cultures
Cultures were grown from fetal Wistar rat brains (E17 pooled rat embryos from one litter) in neurobasal medium supplemented with B27, glutamine, and penicillin/streptomycin (neurobasal medium; Invitrogen, Paisley, UK) using described methods (29)
. Compounds to be tested were applied before the addition of 3-NP (1 mM) in neurobasal medium to induce neurodegeneration. After 48 h, cell viability (survival) was determined using MTT reduction and absorbances were read at 590 nM.
Statistical analyses
Data were normally distributed and results were expressed as mean ± SE. Statistical analysis was performed using GraphPad Prism 4.01 software (GraphPad Software Inc., San Diego, CA, USA) and a 1-way ANOVA with a Dunnet post-test. Significance was determined as P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A). There was some initial loss of body weight due to surgery and pump insertion alone, as seen by PBS-administered, sham-operated animals (Fig. 1A
); however, weight gains were re-established after 2 days. Rats pretreated with the C5aR antagonist PMX205 had a significantly reduced mean body weight loss compared to untreated 3-NP rats from days 2 to 7 (P<0.01; Fig. 1A, B
). Day 7 body weight loss was also significantly reduced in PMX53 pretreated rats and rats treated 2 days after 3-NP administration with PMX205 compared with untreated 3-NP rats (P<0.05; Fig. 1B
). Infliximab or ibuprofen pretreatment resulted in no significant reduction in body weight loss vs. untreated 3-NP rats (P>0.05; Fig. 1B
).
|
The amount of food consumed by rats was also greatly reduced in 3-NP-administered animals. Sham-operated, PBS-administered animals consumed an average of 22.9 ± 0.5 g of food in the 24 h period leading up to euthanasia on day 7 (n=6; data not shown), whereas untreated 3-NP-administered animals consumed 2.9 ± 0.6 g of food (n=28; Fig. 1C
). Rats pretreated with either PMX53 or PMX205, or post-treated with PMX205, had significantly increased food intake compared with untreated 3-NP rats (P<0.05; Fig. 1C
), although not to the extent of sham-operated animals. Infliximab or ibuprofen pretreatment resulted in no significant improvement in food intake compared to untreated 3-NP rats (P>0.05; Fig. 1C
).
3-NP intoxication resulted in progressive behavioral changes and motor deficits that appeared 5 days after administration. Typical deficits involved gait disturbances and ataxia, hind-limb muscle weakness (dystonia), and recumbency. These parameters were scored by a blinded observer using an established scale (24)
. Rats treated with the C5aR antagonists (PMX53 pretreatment, and PMX205 pre- and post-treatment) had significantly reduced neurological deficits compared to untreated 3-NP rats (P<0.05; Fig. 1D
). Rats treated with either infliximab or ibuprofen showed no significant reduction in neurological scores (P>0.05; Fig. 1D
).
C5aR inhibition reduces 3-NP-induced striatal degeneration
Administration of 3-NP to rats resulted in the formation of extensive rounded lesions throughout the striatum at 7 days (Fig. 2
A–C). Lesions consisted of areas of neuronal damage with varying degrees of inflammatory response including hemorrhage, perivascular edema, and neutrophil infiltration (Fig. 2D-K
). Damaged neurons had pyknotic nuclei, and neutrophil infiltration was confirmed using a neutrophil-specific esterase stain (Fig. 2L
). Sham-operated, PBS-administered animals had no detectable histological lesions (Fig. 2)
. Rats treated with C5aR antagonists had markedly fewer severe lesions (Fig. 2)
.
|
All lesions were examined in a blinded fashion and scored on the basis of distinct pathological features (neutrophil infiltration, hemorrhage, perivascular edema, and extent of striatal damage). Lesion sizes were also quantified using computerized software. Pretreatment with either PMX53 or PMX205, or post-treated with PMX205, had significantly reduced lesion sizes and reduced lesion pathology compared to untreated 3-NP rats (P<0.05; Fig. 3
A, B). Rats pretreated with infliximab or ibuprofen had no significant effect on lesion pathology (P>0.05; Fig. 3A, B
).
|
C5aR inhibition reduces 3-NP-induced striatal apoptosis and astrocytosis
Striatal sections were also examined for the degree of apoptosis using the TUNEL method. PBS-administered, sham-operated animals had no detectable apoptotic cells in the striatum (Fig. 4
A). In contrast, 3-NP-administered, untreated animals demonstrated numerous distinct apoptotic cells in the striatal lesions (Fig. 4B
, Fig. 5
). Cells at the edge of the lesion showed the highest degree of apoptosis (Fig. 4B
). Rats treated with the C5aR antagonists had significantly fewer apoptotic cells compared to untreated 3-NP rats (P<0.05), which was also associated with decreased lesion size (Fig. 4C
, Fig. 5
). Rats treated with ibuprofen and infliximab showed a similar degree of apoptotic cells compared to untreated 3-NP rats (Fig. 5)
.
|
|
Astrocyte involvement was examined by immunostaining for glial fibrillary acidic protein (GFAP) in striatal lesions. Staining for GFAP in sham-operated, PBS-administered animals showed a relatively low degree of staining, highlighting astrocytes in their quiescent form (Fig. 4D
). Staining in untreated 3-NP lesioned animals showed a marked up-regulation of GFAP staining on reactive astrocytes around the edge of the striatal lesion (Fig. 4E
). In contrast, PMX205-treated animals demonstrated decreased GFAP staining compared to untreated 3-NP animals, which was associated with decreased lesion formation in these C5aR antagonist-treated animals (Fig. 4F
).
3-NP-induces striatal up-regulation of complement factors
At the conclusion of the experiment on day 7, distribution of the complement factors C3 and C9, as well as complement receptors C5aR and C5L2 in the striatum, were examined by immunohistochemistry. In animals that had received treatment with 3-NP alone, C5aR-, C3-, and C9-positive cells were found to be present on the edge of the 3-NP-induced striatal lesion (Fig. 6
). While C3 and C9 appeared to be confined to the cell body of these cells, C5aRs were present on the cell membrane extending along dendritic processes (Fig. 6C
). In contrast, in 3-NP-administered animals treated with the C5aR antagonist PMX205, C5aR-, C3-, and C9-positive cells were not found to be up-regulated in the striatum. PBS-administered, sham-operated animals also did not display any significant staining of C5aR, C3, or C9. The numbers of C5L2-positive cells seen in the striatum were found to be relatively low regardless of the treatment the animals had received.
|
3-NP-administration results in progressive striatal pathology from day 5 to 7
To determine the onset of striatal pathology after administration of 3-NP, sections from animals that had received 3-NP for 1–7 days were examined for the degree of neutrophil infiltration, hemorrhage, and lesion size. It was found that striatal sections were devoid of any pathology from days 1 to 4 (Table 1
). Striatal lesions began appearing from day 5, and became progressively worse until day 7 (Table 1)
. Hemorrhage within the striatal lesion accompanied the onset and time course severity of the lesion appearance (Table 1)
. However, neutrophil infiltration within the striatal lesions did not appear until day 6, 1 day after the first lesion appearance (Table 1)
. In correlation with the onset of striatal pathology, behavioral deficits began appearing from day 5, and became progressively worse until day 7 (Table 1)
.
|
To determine the time course of expression of C5aRs, C3, C9, and GFAP expression in the striatum during the development of the striatal lesion, sections from the CNS of animals that had received 3-NP for 1–7 days were immunolabeled for C5aRs, C3, C9, and GFAP. Substantial increases in C5aR-, C3-, and C9 immunolabeling only appeared once a lesion had developed, that is, between days 5 and 7 of 3-NP intoxication (Table 1)
. Similar to our earlier finding, these cells were mainly located along the edges of the lesion, although some cells were seen in the lesioned tissue itself. In contrast, animals that had received 3-NP during days 1–4 had very low numbers of C5aR-, C3-, and C9-positive cells in the striatum (Table 1)
. Similarly, the appearance of reactive astrocytes seemed to be dependent on lesion formation, with up-regulation of GFAP-staining observed only between 5 and 7 days of 3-NP administration.
C5aR antagonists are detected in brain after systemic administration
To determine whether the C5aR antagonists can cross the blood-brain barrier, we performed a preliminary investigation where plasma levels and brain tissue levels of the drug were determined in the same animal 20 min after an i.v. dose of 3 mg/kg of PMX53 or PMX205. The tissue levels of the C5aR antagonist PMX53 in the brain were 37.1 ± 14.7 ng/gram (n=4), which represented 1.5% of circulating plasma levels at the time of brain removal (plasma levels: 2.5±0.24 µg/ml, 20 min after PMX53 administration). CNS levels of PMX205 were 31.8 ± 19.5 ng/gram (n=6), which represented 3.1% of circulating plasma levels at the time of brain removal (plasma levels: 1.03±0.15 µg/ml, 20 min after PMX205 administration). These results indicate that the C5aR antagonists are able to cross the intact blood-brain barrier from the circulation.
C5aR inhibition does not prevent 3-NP-induced neurodegeneration on striatal neuron cultures
The ability of the C5aR antagonists to reduce 3-NP-induced striatal pathology in vivo prompted us to determine effects in an in vitro system. Administering 3-NP (1 mM) to primary striatal neuron cultures resulted in prominent cell death as measured by the MTT assay (A590 
40% of untreated culures). Administration of the C5aR antagonists PMX53 and PMX205 to the culture before 3-NP addition had no effect on 3-NP-induced cell death (data not shown). Administration of recombinant mouse C5a, ibuprofen, or infliximab were also without effect on neuronal survival (data not shown).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 9)
. In this study we utilized the orally active cyclic peptide C5aR antagonists AcF-[OP(D-Cha)WR] (PMX53) and hydrocinnamate-[OP(D-Cha)WR] (PMX205), developed in our laboratories, in a rat model of 3-NP-induced neurodegeneration. This model induces severe striatal damage over several days and is used as an acute model of neurodegeneration, which mimics some of the pathology seen in HD (24)
.
We examined basic indices of animal health, including food consumed and body weight, behavioral signs of intoxication, and a variety of CNS histopathological markers. We found that the C5aR antagonists significantly reduced body weight loss, anorexia, and the behavioral and motor deficits associated with intoxication. Postmortem analysis showed distinct striatal lesions in affected rats with marked apoptosis and neutrophil infiltration, a hallmark feature of these lesions (30
, 31)
. Rats treated with C5a antagonists had significantly reduced lesion sizes, associated apoptosis, and neutrophil infiltration. Immunohistochemical analysis of the degenerating striatum demonstrated marked deposition of the complement activation products C3 and C9, and up-regulation of C5aRs on CNS cells at the edges of established and developing lesions. Taken together, these results indicate that C5a plays a role the development of the striatal lesions caused by 3-NP intoxication.
Analysis of the time course of events revealed that despite earlier weight loss there was no striatal lesion evident until day 5, with progressively larger lesions occurring until day 7, when the experiment was terminated. The histological appearance of the striatal lesion occurred at the same time as the onset of neurological symptoms, which provides the underlying structural basis for the neurological deficits. The results clearly demonstrated that complement deposition occurred once the lesion was detectable on day 5. As the lesion developed in size, activation of complement became more intense and was clearly concentrated on the edges of the established lesion. C5aR up-regulation was particularly evident on neurons, but was also observed on other CNS cell types (presumably astrocytes). These results suggest that complement activation by 3-NP in the brain is a relatively slow process in this model. As such, it is possible that blood-brain barrier disruption after chronic 3-NP administration (30)
results in local hemorrhage, and the accompanying activation of complement and accumulation of immune factors not normally present in the CNS which may lead to the onset of striatal degeneration. Indeed, neuronal complement activation is seen after intracerebral hemorrhage in rats (32)
. Further damage likely occurs when neutrophils infiltrating to the damaged striatum are activated by locally produced C5a to induce the release of tissue destructive enzymes (33
, 34)
. Although the precise sequence of events governing how C5a orchestrates the pathology are not yet known, the global reductions in multiple indicators of striatal pathology and associated clinical sequelae by the C5aR antagonists provide compelling evidence that C5a plays a crucial proinflammatory role in this model.
The lack of effect of C5aR antagonists on 3-NP toxicity on cultured neurons indicates that there may be no direct action of the C5aR antagonists in preventing 3-NP cell toxicity, and that the effect of C5a inhibition is due instead to secondary events occurring in the in vivo model. However, different rat strains were used in the in vitro and in vivo models, and this may account for part of the differences seen in the two models. This study also has potential implications for the treatment of other neurological traumas where cell stress (due to anoxia, stroke, or chemical/mechanical trauma) induces complement activation leading to secondary damage. PMX53 was recently shown to reduce cryo-induced acute brain injury in mice (23)
.
In contrast to the other complement components examined, there was little change in C5L2 immunostaining in the striatum over the course of the study. C5L2 is a newly described complement anaphylatoxin receptor whose precise role is not yet fully elucidated (35)
. C5L2 expression has been demonstrated on neurons and astrocytes, and may exert anti-inflammatory functions (36)
. This receptor has also been shown to be down-regulated on neutrophils in experimental and human sepsis in association with increased C5a generation (37)
. The fact that C5L2 immunostaining in the brain appeared to be unchanged after 3-NP administration despite pronounced immunostaining for the other complement factors (C3, C5aR, and C9) is notable. Our results suggest a limited role for C5L2 receptors in this model, but more studies are required to determine the precise role of the C5L2 receptor in neurodegenerative pathology.
C5aR inhibition by administration of the C5aR antagonists almost completely blocked apoptosis in the degenerating striatum. Neuronal C5aR activation has been suggested to be involved in the induction of neuronal apoptosis (10
, 38
, 39)
, although this remains controversial (40
41
42)
. The relationship between C5aR up-regulation, neuronal apoptosis, lesion size and their down-regulation by C5aR antagonists in 3NP rats may be causally related, but more studies are required to determine whether this is the case. Nonetheless, inhibition of C5aRs resulted in a decrease in apoptotic cells, thereby reducing the degree of neurodegeneration.
In this study we used two specific C5aR antagonists: PMX53 and PMX205. PMX53 has been used successfully in various animal models of complement-dependent disease by a number of investigators (16
17
18
19
20
21
22
23)
. We recently reported that the closely related analog PMX205 (14)
is much more potent than PMX53 in a rat model of inflammatory bowel disease when administered orally (15)
and have performed a preliminary comparison of the two drugs in the present study. We observed more consistent therapeutic effects of PMX205 in this study and have hypothesized that the increased lipophilicity of PMX205 over PMX53 may be partly responsible for this. To determine the facility of the drugs to cross the blood-brain barrier, we injected untreated rats with both drugs and determined the plasma and brain levels at the same time point. There was a higher fraction of PMX205 over PMX53 found in the brain relative to plasma levels, and this is consistent with an improved permeation of PMX205 into the brain compared to PMX53. It is possible, however, that the detected brain levels of the antagonists are due to residual intravascular blood not completely washed out in the experiments. More extensive studies are therefore required to fully elucidate the neurological fate of the C5aR antagonists after systemic and oral administration. Since 3-NP toxicity itself is associated with increased permeability of the blood-brain barrier (30)
, the improved activity of PMX205 in these studies requires further investigation. We also administered PMX205 48 h after 3-NP, and found beneficial effects similar to those seen when the drug was preadministered 48 h prior to 3-NP. These experiments were performed prior to the time course studies of striatal lesion development and before the detection of complement activation in the striatum, so the clinical significance of these findings is unclear.
The cyclooxygenase enzymes have been implicated in the pathogenesis of numerous neurological diseases such as Alzheimer’s disease (43)
, and cyclooxygenase-2 enzymes have been shown to be up-regulated in 3-NP intoxicated mice (44)
. In our study, inhibiting cyclooxygenase enzymes with the nonsteroidal anti-inflammatory agent ibuprofen failed to show any overall statistically significant effects, indicating no major pathogenic role for cyclooxygenase enzymes in this model. However, there were marginal changes with ibuprofen treatment for several of the disease parameters, indicating that there may be some minor degree of overall therapeutic activity; further investigation into this is warranted. In addition, the TNF- inhibitor infliximab did not show any protective effects in this model despite being highly protective in a variety of other rat inflammatory disease models (18)
. Future studies should also characterize the effect of further inflammatory pathways by the use of other conventional anti-inflammatories, such as corticosteroids.
In summary, these studies provide the first evidence of a neuroprotective role for C5aR antagonists in an animal model of neurodegeneration. The therapeutic effects of these drugs in this model suggest that inhibitors of C5a function could be beneficial in other models of neurodegeneration. The present study also suggests that neuroinflammation in the form of complement activation and C5a generation plays a deleterious role in 3-NP-induced striatal degeneration, and this result may help direct future investigations into HD. The high efficacy and oral activity of the C5aR antagonists described in this study may indicate a potential future role in the treatment of a variety of human neurodegenerative diseases.
|
ACKNOWLEDGMENTS |
|---|
Received for publication January 24, 2006. Accepted for publication February 27, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. M. Woodruff, K. J. Costantini, J. W. Crane, J. D. Atkin, P. N. Monk, S. M. Taylor, and P. G. Noakes The Complement Factor C5a Contributes to Pathology in a Rat Model of Amyotrophic Lateral Sclerosis J. Immunol., December 15, 2008; 181(12): 8727 - 8734. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
