| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online April 27, 2001 as doi:10.1096/fj.00-0540fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Cell and Neurobiology, Institute of Anatomy,
Neuroscience Research Center, Medical Faculty (Charité), Humboldt-University Berlin, 10098 Berlin; and
Department of Biochemistry and Tumor Biology, Clinic of Obstetrics and Gynecology (OE 6410), Oststadtkrankenhaus, Medical School Hannover, 30659 Hannover, Germany
2Correspondence: Department of Cell and Neurobiology, Institute of Anatomy, Medical Faculty (Charité), Humboldt-University Berlin, Philippstr. 12, 10115 Berlin, Germany. E-mail: oliver.ullrich{at}charite.de
SPECIFIC AIMS
In neuroinflammation, activated microglial cells release large amounts of oxygen free radicals, which might contribute to severe cell damage and death to the microglial cell itself. Conversely, microglial cells have been shown to withstand this cytotoxic action of free radicals. Here we revealed an intracellular mechanism that apparently enables microglial cells to cope with such oxidative challenges.
PRINCIPAL FINDINGS
1. Enhanced protein degradation in activated BV-2 microglial cells
In a first set of experiments, the intracellular degradation of
metabolically radiolabeled endogenous proteins was measured in cellular
or nuclear lysates of tumor necrosis factor 
(TNF-) -treated vs.
resting BV-2 microglial cells in a period of up to 24 h. TNF-
release in the course of neuroinflammation is known to induce an
activated/amoeboid state of BV-2 microglial cells. Whereas the
percentage of protein degradation in resting BV-2 microglial cells was
8.3 ± 0.7% after 24 h, the intracellular protein
degradation was distinctly elevated to 14.0 ± 1.3% in
TNF--treated BV-2 microglial cells. This was accompanied by an even
more distinct degradation rate in isolated nuclei from TNF--treated
BV-2 cells (21.4±1.5%), suggesting that the elevated degradation rate
in total cell lysates after TNF- treatment might reflect mainly the
very high nuclear protein degradation. Inhibition of the microglial
proteasome by the selective proteasome inhibitor MG-132 almost
completely abrogated the enhanced protein degradation after TNF-
treatment in cells and completely in nuclear lysates.
2. Enhanced proteasomal proteolytic activity in activated
microglial cells
Since the proteasome is able to recognize and degrade specifically
oxidatively damaged, an enhanced intracellular protein degradation
could be the consequence of either a higher degree of TNF--induced
oxidative protein damage or an up-regulation of the proteasome
activity. To distinguish between these possibilities, we measured the
degradation of an exogenous [3H]-radiolabeled
native and standardized oxidatively damaged model protein in lysates
from resting or activated BV-2 microglial cells. The exogenous
oxidatively damaged model protein was more efficiently degraded in
TNF--treated, activated BV-2 microglial cells than in resting
BV-2 microglial cells (12.4±1.4% vs. 8.2±0.9%). Experiments
with the specific proteasome inhibitor lactacystine revealed that the
oxidation-specific and stimulatable part of the total proteolytic
activity toward oxidatively damaged histones was a proteasomal
proteolytic activity. Therefore, we conclude that the microglial
proteasome is able to recognize and degrade oxidatively damaged
proteins during the activated state at a significantly higher rate than
during the resting state.
3. The enhanced proteasomal proteolytic activity is dependent on
PARP activity
We measured the proteasome activity directly in resting or
activated BV-2 microglial cells as the lactacystin-sensitive
degradation of the proteasomal fluorogenic peptide substrate
suc-LLVY-MCA. We discovered an 1.6-fold higher
suc-LLVY-MCA-degrading activity in the total cell lysate of
TNF--treated BV-2 cells, whereas this proteolytic activity was
3.9-fold higher in the nucleus after TNF- treatment. This
TNF--stimulatable suc-LLVY-MCA degrading activity could be
completely abolished by lactacystin, indicating that this proteolytic
activity is based on the proteasome function.
Evidence from the endogenous protein degradation, from the degradation
of exogenous oxidatively modified proteins, and from the endogenous
lactacystin-sensitive protease activity revealed an enhanced proteasome
activity in activated BV-2 microglial cells in comparison to resting
cells. Since previous studies demonstrated an activating interaction
between the proteasome and the functional active nuclear enzyme
poly-ADP-ribose polymerase (PARP) in vitro, we tested the effect of the
PARP inhibitor 3-aminobenzamide (3-ABA) on the endogenous proteasome
activity in cellular and nuclear lysates of resting or TNF--treated
BV-2 microglial cells. Coincubation with 3-ABA reduced the
TNF--induced higher suc-LLVY-MCA degradation by fivefold in
nuclei, indicating an involvement of the functional active PARP.
4. Protein–protein interaction between PARP and proteasome
To investigate the previously demonstrated functional in vitro
interaction of the PARP with the proteasome in living BV-2 microglial
cells, both resting and activated, we performed immunoprecipitation
experiments of the PARP in nuclear lysates of BV-2 microglial cells.
Freshly isolated nuclei from resting or TNF--treated, activated BV-2
microglial cells were lysed and PARP protein was precipitated by an
antibody directed against the DNA binding domain of the PARP. These
immunoprecipitations contributed to a coprecipitation of proteasome
subunits and proteasomal proteolytic activity, but only under the
conditions of enzymatic active PARP. After PARP inhibition by 3-ABA,
PARP protein was precipitated, but without coprecipitation of
proteasome subunits or activity in the precipitates, indicating that
PARP–proteasome interaction requires the functional active PARP.
Consequently, the specificity of the central function of the PARP was
addressed in the following experiment.
5. PARP protein is specifically involved in proteasome
up-regulation
We constructed an antisense PARP-pTracerCMV2 vector and
established a stably transfected BV-2 microglial cell line carrying the
antisense PARP-pTracerCMV2 vector as well as the pTracerCMV2 vector
alone. After transformation of these stable transfected BV-2 microglial
into the activated state by TNF-, total protein degradation in
metabolically prelabeled cells (Fig. 1A
) and proteasome activity in the isolated nuclei were
measured (Fig. 1B
) in comparison to resting cells. The
antisense effect of the asPARP vector vs. control vector-transfected
cells is demonstrated in Fig. 1C
. In comparison with the
pTracerCMV2 vector carrying control BV-2 cells, asPARP transfected BV-2
cells failed to respond with an elevated protein degradation after
TNF- treatment (Fig. 1A
). The endogenous cellular and
nuclear proteasome activity in resting BV-2 microglial cells was
reduced twofold in the asPARP transfectants vs. control cells and
nuclear proteasome activity was twofold less stimulatable after TNF-
administration (Fig. 1B
). These results indicate that
the PARP protein is specifically involved in the proteasome
up-regulation in TNF--activated BV-2 microglial cells.
|
6. Proteasome activity protects activated BV-2 micoglial cells from
oxidative damage
The selective recognition and degradation of damaged cellular
proteins are well known and widely investigated biological functions of
the ATP- and ubiquitin-independent proteasomal pathway. Therefore, an
increased proteasome activity in activated phagocytosing microglial
cells, which produces large amounts of oxygen radicals, might be
important in self-protection against this oxidative damage or in
regulating the degradation of incorporated material after phagocytosis.
To test this hypothesis, the total protein carbonyl content as a marker
of the endogenous protein oxidation was studied in the presence and
absence of the selective and reversible proteasome inhibitor MG-132.
This substance induced 80% proteasome inhibition in the applied
concentration and time range without reducing cell viability (data not
shown). TNF--treated BV-2 microglial cells showed an increased
oxidation of endogenous proteins after proteasome inhibition. No
difference in the accumulation of protein carbonyls could be observed
in activated BV-2 cells with active proteasome vs. resting BV-2 cells.
Inhibition of the PARP by 3-ABA also contributed to increased carbonyl
content after TNF- treatment, though not as high as during
proteasome inhibition by MG-132. This suggests that the constitutive
proteasome activity might exert a significant prevention from oxidative
cell damage. Furthermore, BV-2 cells with inhibited proteasome activity
demonstrated a drastically reduced viability after TNF- treatment
(10% of the cells without proteasome inhibition) and exhibited
characteristics of apoptotic cell death like nuclear condensation and
fragmentation. Similarly, inhibition of the PARP resulted in a loss of
viability, but not to the extent obtained during proteasome inhibition.
CONCLUSIONS
Activation of microglial cells is accompanied by a strong
respiratory burst, during which large amounts of oxygen free radicals
are produced and released. Once activated, microglia can promote
neuronal injury through the release of neurotoxins like cytokines,
oxidized lipids, and oxygen free radicals. However, it remained unclear
why this release does not result in severe toxicity to the microglia
itself. If microglia are programmed to remain viable during their
conversion to the activated state, during which they produce and
release large amounts of free radicals, they should harbor cellular
mechanisms rendering resistance to free radical toxicity. Here we
demonstrated that the proteasome plays a crucial role in microglial
self-protection. It is involved in the enhanced protein turnover and
degradation of oxidatively modified proteins after TNF--induced
microglial cell activation and its activity is regulated by the
interaction with the functional active PARP, which protects activated
microglia from protein oxidation and cell death.
Our results demonstrated that inhibition of PARP contributed to microglial cell damage and death specifically of activated, but not of resting, cells by the impairment of an antioxidative metabolic pathway. Therefore, pharmacological inhibition of the PARP should specifically kill activated and not resting microglial cells. This might be beneficial, particularly in neuroinflammatory diseases where activated microglia are involved as a pathological or copathological factor. The complex function of microglia during neuroinflammation, however, requires a detailed understanding of intracellular pathways for inducing or rendering a specific functional state of these cells in order to provide a basis for targeted intervention on microglial reactions during neuroinflammation.
|
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.00-0540fje ; to
cite this article, use FASEB J. (April 27, 2001) 10.1096/fj.00-0540fje 
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
