| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Research Communications |
Department of Neurobiology, The Weizmann Institute of Science, 76100 Rehovot, Israel
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: thrombin • matrix metalloproteases • CNS • PNS
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
One of the proteases known to influence the nervous system is thrombin, a key enzyme in blood coagulation (9) . Thrombin acts as a potent mitogen that modulates the morphology of astrocytes 10, 11) and can induce the secretion of nerve growth factor (NGF) by astrocytes (12) . It inactivates acidic fibroblast growth factor, known to promote neuritic outgrowth and astrocyte proliferation 13-15) . It inhibits outgrowth from neuritic neuronal cells due to interaction with the thrombin receptor PAR-1 16, 17) , which is proteolytically activated by thrombin (7) . Thrombin at high concentrations was shown to induce apoptotic cell death in astrocytes and neurons cultured under normal conditions, whereas at moderate concentrations it protected those cells from a variety of metabolic insults 18-20) . Both effects were shown to be due to the activation of the PAR-1 receptor. The processes of neuroprotection and apoptotic cell death display similar characteristics in the signal transduction pathways 19, 21, 22) , and it was suggested that different concentrations of thrombin might result in different activation levels of the same pathway (21) .
Another group of proteases involved in wound healing is the family of
matrix metalloproteases (MMPs, collagenases). Two of these, gelatinase
A (MMP-2, 72 kDa) and gelatinase B (MMP-9, 92 kDa), belong to the
subfamily of gelatinases. MMPs are largely responsible for the
degradation of extracellular matrix (ECM) components such as collagen
and proteoglycans in several normal and pathological processes,
including tissue remodeling (23)
, and MMP genes are among
the most abundant of those expressed by cells in inflammatory lesions
(24)
. Remodeling of the ECM by proteolytic activity is
known to be crucial for growth-cone motility (8)
. Whereas
thrombin is known to produce the matrix component fibrin 9, 25, 26)
, MMPs are responsible for matrix degradation. Fibrin
fibronectin-containing matrix promotes the growth of cultured septal
and hippocampal neurons from newborn rats (27)
. The MMPs
are activated by the fibrin-degrading plasmin, which is itself
activated by plasminogen activators (PA) (24)
. Cultured
neurons from the PNS release at least two proteases capable of
degrading ECM components, a calcium-dependent MMP and a PA
(28)
. In addition, NGF-induced expression of gelatinase A
by dorsal root ganglionic neurons occurs in correlation with the
ability to degrade ECM and to extend neurites (29)
, and
NGF production in astrocytes is induced not only by thrombin but also
by gelatinases (12)
. In contrast to the above findings,
which point to beneficial effects of MMPs in the nervous system,
several studies hint at the involvement of MMPs in processes
detrimental to the brain. Intracerebral injection of bacterial
collagenase as well as of tumor necrosis factor (TNF) -
induces
gelatinase B production, which causes delayed opening of the
blood–brain barrier 30, 31)
. Patients with inflammatory
neurological disorders, including multiple sclerosis, show increased
gelatinase B in the cerebrospinal fluid (32)
. Gelatinase B
is also present in the cerebrospinal fluid of mice with experimental
allergic encephalomyelitis, the animal model of multiple sclerosis,
where it cleaves myelin basic protein (33)
.
To investigate the relationship of gelatinase and thrombin activities to regenerative and degenerative processes in the nervous system in vivo, we compared their activities in response to axonal injury in two regenerative nervous systems (rat sciatic nerve and fish optic nerve) with a nonregenerative nervous system (rat optic nerve). We show that the induction of gelatinase activity shortly after injury is comparable in the three nerve types. In contrast, the postinjury induction of thrombin-like activity in the nonregenerative optic nerve differs clearly from that in both regenerative nerves in a way that could explain its detrimental and beneficial effects, respectively, on regeneration.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Crush injury
Rats were anesthetized with 10 mg/kg xylazine (Vitamed, Israel)
and 50 mg/kg ketamine (Fort Dodge Laboratories, Fort Dodge, Iowa). For
the optic nerve crush, a lateral canthotomy was performed in the right
eye under a binocular operating microscope as described previously
(34)
. The conjunctiva was incised laterally to the cornea,
the retractor bulbi muscle was separated, and the optic nerve was
exposed. The dura was left intact. For the sciatic nerve crush, a small
incision was made in the thigh to expose the nerve. The exposed optic
or sciatic nerve was crushed by calibrated forceps for 30 s. After
the sciatic nerve crush, the skin was sutured. The rats were allowed to
recover.
Fish were deeply anesthetized with 0.05% 3-aminobenzoic acid ethyl ester (Sigma, St. Louis, Mo.). As in the rat, the conjunctiva was incised laterally to the cornea and the exposed optic nerve was crushed intraorbitally with forceps for 30 s. The fish were then returned to their tanks.
Preparation of conditioned media
At specified times after crush injury, the animals were again
anesthetized and their crushed nerves were dissected out, cleaned with
Kimwipes, and washed in Dulbecco's modified Eagle's medium (DMEM)
without phenol red. Each of the three types of nerves was pooled and
incubated separately for 1.5 h at room temperature in DMEM without
phenol red. The nerves were then removed and the three resulting media,
now termed `conditioned media', were collected, centrifuged at
15000 x g for 15 min in an Eppendorf centrifuge at
4°C, and stored at -70°C. Before use, the conditioned media were
kept on ice to limit proteolysis. Protein concentrations were
determined by Bradford analysis (Bio-Rad, Hercules, Calif.). For some
experiments, nerves were separated into proximal and distal segments
before incubation. As controls, conditioned media were similarly
prepared from uninjured nerves.
Preparation of nerve extracts
As in preparation of the conditioned media, the animals were
reanesthetized at specified times after crush injury. The crushed
nerves were dissected from their sheaths and some were separated, as
before, into proximal and distal segments. The pooled nerves or nerve
segments were then homogenized and left overnight at 4°C in
extraction buffer containing 0.5% Triton X-100, 0.1 M Tris-HCl, pH
8.0. After centrifugation at 15,000 x g for 15 min at
4°C, the protein concentrations of the supernatants were determined
by Bradford analysis (Bio-Rad). Nerve extracts were stored at -70°C
and prior to the activity assays were kept on ice to limit proteolysis.
As controls, nerve extracts were similarly prepared from uninjured
nerves.
Thrombin activity assay
An assay for thrombin activity was designed according to a
previously described procedure (35)
. Samples of the
conditioned media, each containing 5 µg of total protein, were
incubated at 37°C in a flat-bottomed, 96-well plate with 50 mM
Tris-HCl, 5 mM EDTA, pH 8.4, and 150 µM of
N-p-Tosyl-Gly-Pro-Arg-p-nitroanilidine (Chromozym-TH;
Sigma), a specific substrate for thrombin, in DMEM without phenol red
in a final volume of 200 µl. In some experiments, the thrombin
inhibitor hirudin (Sigma) or thromstop (American Diagnostica,
Greenwich, Conn.) was added. Optical density (OD) at 405 nm was
measured at the times specified. Wells containing all of the above
except for the conditioned media were used as blanks, and signals in
the tested samples were expressed as the difference between the
measured OD and the OD recorded at the beginning of the assay. Only
those measurements obtained when enzymatic activity was in the linear
range are shown. In some cases, pure thrombin from bovine plasma
(Sigma) was assayed in order to quantify the signals of the assay.
Thrombin activation by ecarin
Samples of the conditioned media, each containing 5 µg of
total protein, were incubated for 2 h at 37°C in a
flat-bottomed, 96-well plate with 50 mM Tris-HCl, 5 mM EDTA, pH 8.4, in
DMEM without phenol red, in a final volume of 100 µl with 0.1 U or 1
U of ecarin (American Diagnostica), a specific prothrombin activator
derived from snake venom. Controls were treated in the same way, but
without ecarin. Thrombin-like activity was then assayed as described
above. Most of the existing prothrombin was activated by 0.1 U of
ecarin, as increasing the concentration to 1 U was found to have no
further effect on the signal.
Gelatin zymography
Conditioned media or nerve extracts containing equal amounts of
protein were electrophoresed through an 8% sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel (Bio-Rad)
that was polymerized in the presence of 1 mg/ml gelatin. To produce a
linear polyacrylamide gradient gel, a 7% and a 15% gel solution—each
containing 1 mg/ml gelatin—were mixed by a gradient mixer and poured
with a peristaltic pump at a speed of 2 ml/min, as described
(36)
. The gels were washed once for 30 min in 2.5% Triton
X-100 to remove the SDS, and once for 30 min in the reaction buffer (50
mM Tris-HCl, 200 mM NaCl, 10 mM CaCl2, pH 7.5). The gels
were then incubated in fresh reaction buffer at 37°C for 3 days.
Gelatinolytic activity was visualized by staining of the gels with
0.5% Coomassie brilliant blue.
Statistics
The significance of differences was calculated using the Welch
alternate t test. This test assumes Gaussian populations,
but does not assume that the compared populations have equal standard
deviations, and is therefore more conservative than the conventional
unpaired Student's t test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
The sciatic nerve also showed an increase in thrombin-like activity 1 day after the injury, but this elevation was approximately 1/20th that of the optic nerve (Fig. 1A ) and could be simulated by 0.5 x 10-3 U of pure bovine thrombin. Because these signals were very low, we extended the incubation time with Chromozym-TH from 1 to 10 h (Fig. 1A , insert). In both cases, the thrombin-like activity in the sciatic nerve on day 1 postinjury was approximately threefold higher than in the uninjured sciatic nerve (P<0.001) and had returned to baseline by 4 days after injury. Thus, both the sciatic and the optic nerve showed an increase in thrombin-like activity 1 day after injury, but the increase in the optic nerve was approximately 20-fold higher than in the sciatic nerve.
Specific inhibition of thrombin-like activity in the rat
To further verify the specificity of the measured thrombin-like
activity, we analyzed it in the presence of the synthetic thrombin
inhibitor thromstop (Fig. 2
) or hirudin (not shown). At the different time points examined,
similar dose-dependent patterns of inhibition were observed in the
optic and the sciatic nerves: the presence of thromstop at a
concentration of 0.1 µM inhibited activity by approximately 60%, and
at a concentration of 1 µM by about 90%.
|
Thrombin-like activity is potentially associated with retrograde
degeneration
To find out whether the elevated thrombin-like activity in the
injured rat optic nerve on day 1 postinjury is associated with
retrograde or anterograde degeneration, some of the nerves were divided
into a distal segment (between the injury site and the optic chiasma)
and a proximal segment (including the site of injury and the part of
the nerve between the injury site and the optic disk). Conditioned
media from the pooled proximal segments and the pooled distal segments
were collected separately and their thrombin activities were assayed.
In all cases examined, thrombin activity was found mainly in the
proximal segments (Fig. 3
).
|
Increase of thrombin-like activity by prothrombin activation
One possible source for the observed activity could be the
activation of prothrombin, known to be expressed in the nervous system
(37)
. To compare the amount of prothrombin (a reservoir
for thrombin-like activity) in the rat optic or sciatic nerves before
and after injury, we measured prothrombin levels in the various
conditioned media. After the addition of ecarin, which converts
prothrombin to thrombin, thrombin-like activity was assayed as before.
In the uninjured optic nerve (Fig. 4A ), incubation with 0.1 U of ecarin did not evoke a thrombin activity signal. In the injured optic nerve, however, the high signal observed 1 day after injury was further increased approximately threefold by ecarin, suggesting that the amount of thrombin-like activity in the nerve after the injury had corresponded to about 30% of the potentially available activity. In the sciatic nerve (Fig. 4B ), incubation with ecarin increased the signal of thrombin-like activity by more than 10-fold in both injured and uninjured nerves. Thus, regardless of injury, it appears that less than 10% of the potentially available thrombin-like activity in the sciatic nerve exists in the active form.
|
Thrombin-like activity in fish optic nerve
The kinetics of thrombin-like activity was studied in a similar
manner in the regenerating fish optic nerve. Figure 5A
shows the thrombin-like activity measured in fish optic nerve on
different days after injury as well as in uninjured nerves. Samples of
conditioned media were incubated with the chromogenic substrate
Chromozym-TH and the OD values were measured after 4 h. Injury of
the fish optic nerve resulted in an increase in thrombin-like activity
with kinetics that differed from those observed in the rat. The
activity started to increase on day 1, reached a broad peak that lasted
from day 4 to day 7 after the injury, and then declined. The observed
increase was significant when compared to that in uninjured nerves
(P<0.05).
|
Specific inhibition of thrombin-like activity in fish optic nerve
Thrombin-like activity in the fish optic nerve was inhibited in a
dose-dependent manner by the synthetic inhibitor thromstop (Fig. 5B
) or hirudin (not shown), demonstrating its specificity.
Much larger amounts of both inhibitors were needed to achieve
inhibition in the fish nerve than in either of the rat nerves: for
example, to achieve an inhibition of about 60% in the fish nerve it
was necessary to use 100 µM thromstop, a concentration 1000-fold
higher than that needed in the rat. The inhibition pattern in the fish
optic nerve was identical at several examined time points.
Activities of gelatinases in rat optic and sciatic nerves
To examine whether the observed up-regulation of thrombin-like
activity in the injured optic nerve of the rat is unique to thrombin or
merely part of a general, nonspecific up-regulation of protease
activity after injury (which would imply a general difference in
protease regulation between regenerative and nonregenerative nervous
systems), we also measured the postinjury activities of gelatinases. As
mentioned earlier, gelatinases are known to participate in processes
that are related to both regeneration 27-29)
and
degeneration 30-33)
in the nervous system. Conditioned
media and nerve extracts from rat optic and sciatic nerves were
collected at specified times after injury, as well as from uninjured
nerves, and gel zymography was performed. Figure 6
shows representative gels of gelatinase activities in rat optic (Fig. 6A
) and sciatic (Fig 6B
) nerves. In both cases,
gelatinase A (72 kDa) was found to be constitutively expressed and did
not increase after injury. In contrast, gelatinase B (92 kDa) could not
be detected in either of the nerves before injury, but was up-regulated
in both 1 day after injury, after which its activity declined. In the
rat optic nerve, this up-regulation was detectable as early as 6 h
after the injury (Fig. 6A
, lane 2). In some optic nerve
preparations, an additional band below 200 kDa was visible (Fig. 7A
, lane 1), probably representing a complex formed between
a gelatinase and an inhibitor. Apart from the existence of this
complex, the kinetics of the rat sciatic and optic nerves were similar.
Identical results were obtained when nerve extracts were used instead
of conditioned media (not shown).
|
|
Gelatinase activity is potentially associated with retrograde
degeneration
Having found that thrombin-like activity in the rat optic nerve is
potentially associated with retrograde degeneration (Fig. 3)
, we were
interested in establishing which parts of the injured optic and sciatic
nerves are responsible for the observed changes in gelatinase activity.
Accordingly, both nerves were again divided into a proximal segment,
which included the site of injury, and a distal segment. Nerve extracts
were prepared from each segment separately. In both optic and sciatic
nerves, induction of gelatinase B was found to occur only in the
proximal segment (Fig. 7
). Identical results were obtained when conditioned media were used
(not shown).
Gelatinase activities in the fish optic nerve
As with thrombin, we studied the kinetics of gelatinase activities
in response to axonal injury in the regenerating fish optic nerve. As
before, conditioned media and nerve extracts were collected at
specified times after the injury, as well as from uninjured nerves, and
gel zymography was performed. Figure 8
shows representative gels of gelatinase activities in the fish optic
nerve. Two major bands were found at 72 and 92 kDa. Since they were
identical in size and exhibited similar postinjury behavior to that of
the gelatinase activity observed in the rat, they were assumed to be
homologous to rat gelatinases A and B, respectively. As in the rat, the
72 kDa gelatinase was found to be constitutively expressed and
unaffected by the injury, whereas the 92 kDa gelatinase was
up-regulated. Several additional phenomena, not present in the rat,
were observed in the fish. 1) Up-regulation of gelatinase
activity after injury remained high for much longer than in the rat;
for example, baseline levels of the 92 kDa gelatinase were not reached
even by day 7. 2) In uninjured nerves, the ratio between the
gelatinases at 72 and 92 kDa differed from that obtained in the rat,
indicating a greater abundance of the 92 kDa gelatinase in the fish.
3) Additional bands at about 20, 35, 45, 55, and 80 kDa were
up-regulated upon injury. These bands might be a result of
up-regulation of additional gelatinases or of proteolysis of the known
gelatinases where the cleaved products were still active. 4)
Less protein was needed in the fish conditioned medium in order to
achieve a band intensity similar to that of the rat conditioned medium,
indicating that the activity in the former medium is higher, either
because gelatinases are more abundant or because the activities of the
fish gelatinases are higher than those of the rat. Identical results
were obtained when nerve extracts were used instead of conditioned
media (not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
High levels of thrombin activity were previously shown to inhibit neurite outgrowth (16) and to induce apoptotic cell death in neurons and astrocytes 20-22) in vitro. The high levels of thrombin-like activity in the rat optic nerve observed 1 day after injury in our study suggest that the degenerative processes seen after CNS injury might be related to thrombin activity and that thrombin-like activity might contribute to the nonregenerative nature of the optic nerve environment. The marked increase in activity could also override the beneficial effects of thrombin that are expected (from in vitro data) to occur in the CNS, such as the stimulation of NGF secretion by astrocytes (12) . In contrast, a slightly increased thrombin-like activity, such as that seen in the rat sciatic nerve, might not only be below the toxicity threshold but might even be beneficial for the injured nerves. In the PNS, the injured nerve is rapidly invaded by macrophages, which are crucial for regeneration in the nervous system 1, 38-40) . Thrombin is a potent chemoattractant for human monocytes (41) and stimulates the production of monocyte chemotactic proteins 42-44) . In the mammalian CNS, where macrophage recruitment is impaired (38) because of an immune brain barrier (45) , these potential effects of thrombin could be minimized or excluded. In addition, levels of thrombin-like activity in the sciatic nerve could be within the range at which neurons are rescued from cell death resulting from metabolic insults, which likely to be present after nerve injury 20, 21) .
Although it cannot be ruled out, the possibility that the observed thrombin-like activity results from a protein other than thrombin seems very unlikely, an assumption supported both by the cleavage of the specific thrombin substrate Chromozym-TH and the inhibition by two specific thrombin inhibitors. Thrombin activity results from the cleavage of prothrombin, which is known to be expressed in the nervous system (37) . The experiments with ecarin showed that activation of the available prothrombin would increase thrombin activity by more than 10-fold in the injured sciatic nerve, but by only about 3-fold in the injured optic nerve. This suggests that prothrombin activation is under tighter control in the sciatic nerve than in the optic nerve. A prothrombin activator independent of the blood coagulation cascade was recently described, and its involvement after injury was suggested 46, 47) . Although not detected in uninjured brain, it was found to be active in a cell line derived from mammalian CNS (46) . In addition, prothrombin activation has been shown to occur on neuronal surfaces (48) . Our data suggest that different mechanisms of prothrombin activation in the optic and sciatic nerves could account for the high levels of thrombin-like activity observed after injury in the rat optic nerve.
In a recent study of thrombin and prothrombin elevation in rat sciatic nerve (49) , thrombin-generated activity increased significantly from day 1 after nerve crush, reaching a peak on day 3 and returning to basal levels on day 6. Prothrombin could not be detected in uninjured nerves, but the ratio of prothrombin-to-thrombin activity after injury was comparable to the ratio obtained in our study, and led those authors to suggest a tight control of thrombin activation in the sciatic nerve. The different kinetics obtained in that study (reflecting a relatively delayed increase in thrombin activity) and the inability to detect prothrombin in the uninjured sciatic nerve might be attributable to the methodology, as nerve extracts and not conditioned media were used.
Differences in thrombin-like activities between the rat optic and sciatic nerves could also result from changes in the levels of thrombin inhibitors. A known thrombin inhibitor in the nervous system is the glial-derived protease nexin I (PN-I) (50) , which promotes neurite outgrowth 51, 52) and could therefore reverse the negative effects of thrombin-like activity. In the sciatic nerve, PN-I is up-regulated after injury, starting on about day 3 and peaking on day 7 49, 53) . The behavior of PN-I shortly after injury in the optic nerve has not been studied. Nevertheless, our finding of high thrombin-like activity in the optic nerve 1 day after injury may reflect the paucity or absence of thrombin inhibitors, such as PN-I, at this time.
In the fish, the kinetics of thrombin-like activity in the optic nerve differed from those in the rat. As in the rat, the increase in activity started on day 1 but, unlike in the rat, continued and reached a plateau lasting from day 4 to day 7 after injury. As similar peaking in the fish nerve was also observed in the activities of the gelatinases, this elevation pattern may represent a general response of this regenerative nerve to injury. The kinetics observed in the fish optic nerve point to the establishment of a different extracellular milieu, which is generated by the proteases and could account, at least in part, for the regeneration-supportive properties of the neuronal environment. As in the rat, thrombin-like activity in the fish optic nerve was specifically inhibited, in a dose-dependent manner, by the inhibitors thromstop and hirudin. Much larger amounts of these inhibitors were needed in the fish than in the rat, indicating the presence of other substances in the fish conditioned medium that influence the interaction between the inhibitors and the enzyme.
Studies in our laboratory recently demonstrated the presence of factor XIIIa in the nervous tissues of rat and fish, as well as a correlation between the postinjury appearance and activation of the enzyme and the regenerative ability of the tissue (54) . Thrombin is crucial for the regulation of factor XIIIa in vivo, and the differential activities of thrombin might be responsible for the differences in the observed behavior of factor XIIIa.
We found that the thrombin-like activity in the rat optic nerve was associated mainly with the part of the nerve adjacent to and including the site of injury, i.e., the proximal segment. This suggests that the observed injury-induced activity is associated with retrograde degeneration and, accordingly, is evoked either by changes emanating from the neuronal cell bodies or from a non-neuronal cell response at the site of injury. As mentioned earlier, prothrombin mRNA was shown to be expressed by cells of the nervous system (37) . In addition, thrombin has been detected in brain and astroglial cell cultures (55) . Nevertheless, thrombin synthesis could also occur elsewhere, for example, in platelets that adhere to the site of injury. The cellular source responsible for thrombin activity in the nervous system, and particularly after injury, is not yet known.
In contrast to the differences in thrombin-like activity, we found that the activation pattern of gelatinase A or B was similar in all three nerve types. Gelatinase B was up-regulated after crush injury, whereas gelatinase A was not affected. In the rat, up-regulation of gelatinase B (which did not occur in the uninjured nerve) peaked about 1 day after optic or sciatic nerve injury and decreased on about day 4. The fish showed a different proportion between the two gelatinases in the uninjured nerve, as well as a relatively delayed peak of the activity. As already mentioned, this delayed peaking (which is not exclusive to gelatinases) may influence the regenerative properties of the fish optic nerve.
Gelatinases are known to participate in processes that are related to both regeneration 27-29) and degeneration 30-33) in the nervous system. This study shows not only that gelatinases are present in regenerative and nonregenerative white matter but that, surprisingly, the induction patterns of gelatinases A and B in all three nerve types examined are similar. One possible explanation is that rather than being directly related to degeneration or regeneration, the mechanisms of activation or inhibition of gelatinases in degenerating and regenerating systems might differ. Gelatinase activity can be inhibited by tissue inhibitors of matrix metalloproteases (TIMPs). TIMP-1 was shown to be induced in the sciatic nerve on days 1 and 4 postinjury, and immunostaining demonstrated its colocalization with Schwann cells and macrophages (56) . Another possibility is that not only total protease activity, but also the ratio of ECM-degrading to ECM-producing proteases, might play a role in determining the permissiveness of the extracellular milieu to regeneration. This is in line with our finding that the nonregenerative rat optic nerve differed significantly from the regenerative rat sciatic nerve with respect to the amounts of fibrin-producing thrombin, but not of ECM-degrading gelatinases.
As with thrombin-like activity, gelatinase activities in the rat optic
and sciatic nerve were limited to the proximal segment, indicating that
the observed activities are associated not with anterograde
degeneration, but with an active process occurring at the site of
injury or in the proximal part of the nerve. Our findings in the
sciatic nerve are in line with the observations of La Fleur et al.
(56)
. The cellular source of the gelatinases in
vivo remains to be established. However, the inflammation-related
cytokines interleukin (IL) -1, IL-1ß, and TNF- are potent
inducers of gelatinases in cultured rat astrocytes (57)
,
and activated microglia release gelatinase B in vitro
(58)
. Gelatinase activity has also been reported in
Schwann cells and glioma cells 59, 60)
.
In summary, a comparison of the general kinetics of the induced thrombin-like and gelatinase activities in rat and fish showed that activity in the rat peaks early and declines between days 1 and 4 postinjury, whereas in the fish the activity lasts for at least a week. A comparison of activity levels in the optic and sciatic nerves of the rat showed significant differences in thrombin-like activity but similarity in the activities of gelatinases. The induction pattern of the 72 kDa and 92 kDa gelatinases was similar in fish as well. Both the different kinetics and the differential activity levels might affect the nerve's ability to regenerate. This study further highlights the fact that tissue repair and regeneration are closely linked and that substances that generally participate in wound healing establish an extracellular milieu that can lead either to regeneration or its failure.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Abbreviations: CNS, central nervous system; DMEM,
Dulbecco's modified Eagle's medium; ECM, extracellular matrix; IL,
interleukin; MMP, matrix metalloprotease; MMP-2, gelatinase A; MMP-9,
gelatinase B; NGF, nerve growth factor; OD, optical density; PA,
plasminogen activator; PN-I, glial-derived protease nexin I; PNS,
peripheral nervous system; SDS-PAGE, sodium dodecyl
sulfate-polyacrylamide gel electrophoresis; TIMPs, tissue inhibitors of
matrix metalloprotease; TNF, tumor necrosis factor.
Received for publication July 27, 1998.
Revision received October 29, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. Shavit, O. Beilin, A. D. Korczyn, C. Sylantiev, R. Aronovich, V. E. Drory, D. Gurwitz, I. Horresh, R. Bar-Shavit, E. Peles, et al. Thrombin receptor PAR-1 on myelin at the node of Ranvier: a new anatomy and physiology of conduction block Brain, April 1, 2008; 131(4): 1113 - 1122. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. Lino, S. Atanasoski, M. Kvajo, B. Fayard, E. Moreno, H. R. Brenner, U. Suter, and D. Monard Mice Lacking Protease Nexin-1 Show Delayed Structural and Functional Recovery after Sciatic Nerve Crush J. Neurosci., April 4, 2007; 27(14): 3677 - 3685. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Rohatgi, F. Sedehizade, K. G. Reymann, and G. Reiser Protease-Activated Receptors in Neuronal Development, Neurodegeneration, and Neuroprotection: Thrombin as Signaling Molecule in the Brain Neuroscientist, December 1, 2004; 10(6): 501 - 512. [Abstract] [PDF]
|
|
|
|
 |
|
 R. A. Adams, M. Passino, B. D. Sachs, T. Nuriel, and K. Akassoglou Fibrin Mechanisms and Functions in Nervous System Pathology Mol. Interv., June 1, 2004; 4(3): 163 - 176. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
