| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Animal Models |
From the Robert-Koch-Institute,* Berlin, Germany; the Department of Neurology, Georg-August-University, Göttingen, Germany; and the Department of Environmental Sciences, Eidgenossische Technische Hochschule Zürich, Swiss Federal Institute of Technology, Zürich, Switzerland
Abstract
Prion-induced chronic neurodegeneration has a substantial inflammatory component, and the activation of glia cells may play an important role in disease development and progression. However, the functional contribution of cytokines to the development of the gliosis in vivo was never systematically studied. We report here that the expression of interleukin-1ß (IL-1ß), IL-1ß-converting enzyme, and IL-1 receptor type 1 (IL-1RI) is up-regulated in a murine scrapie model. The scrapie-induced gliosis in IL-1RI–/– mice was characterized by an attenuated activation of astrocytes in the asymptomatic stage of the disease and a reduced expression of CXCR3 ligands. Furthermore, the accumulation of the misfolded isoform of the prion protein PrPSc was significantly delayed in the IL-1RI–/– mice. These observations indicate that IL-1 is a driver of the scrapie-associated astrocytosis and possibly the accompanying amyloid deposition. In addition, scrapie-infected IL-1RI-deficient (IL-1RI–/–) mice showed a delayed disease onset and significantly prolonged survival times suggesting that an anti-inflammatory therapeutical approach to suppress astrocyte activation and/or glial IL-1 expression may help to delay disease onset in established prion infections of the central nervous system.
and interleukin (IL)-6 were shown to have no effect on scrapie pathogenesis and length of incubation periods on intracerebral infections.5,12,13
IL-1, often regarded as prototypic inflammatory cytokine, is known to play a detrimental role in acute neurodegeneration after brain injury and stroke14,15
and was suggested to be involved in Alzheimer’s disease pathogenesis.16
In addition, certain polymorphisms in the genes encoding IL1-1 and IL1-1ß are considered to increase the risk for early disease onset in Alzheimer’s disease.17
Previous studies demonstrated that IL-1 is produced by microglia on exposure to PrP106–126 and ß-amyloid in vitro.18,19 Glial overexpression of IL-1 was previously observed in murine scrapie models during preclinical and clinical stages of the disease,5,13,20,21 in brain tissue of mice transgenic for the human amyloid precursor protein (huAPP) and in Alzheimer’s disease.16,22,23 However, despite this accumulated evidence for an involvement of IL-1 in chronic neurodegeneration, the functional biology of IL-1 overexpression in neurodegenerative amyloidoses was never experimentally addressed. Here we investigated the role of IL-1 in the scrapie-associated glia activation.
Materials and Methods
Mouse Strains and Scrapie Infections
IL-1RI–/– mice, which originated from 129Sv x C57/B6 mouse strain crosses, were maintained at the ETH Zurich, Zurich, Switzerland, and were backcrossed for seven generations into the C57/B6 strain. All infections were performed intracerebrally using 20 µl of diluted 10% brain homogenates prepared from scrapie strain 139A (kindly provided by R.H. Kimberlin, Edinburgh, UK)-infected mice as previously described.24 Further infection experiments using first generation wild-type 129Sv x C57/B6 crossbreeds showed no significant survival time differences to similarly infected wild-type C57/B6 mice, indicating that the genetic background has no influence on the disease development in the used combination of scrapie and mouse strain. Mock infections were similarly performed with 10% brain homogenates prepared from uninfected, healthy C57/B6 mice. All infections with a given dilution were performed at the same day from the same brain homogenate aliquot to minimize variations of the infectious doses. The statistical analysis of the survival of the IL-1RI–/– mice and the wild-type controls infected with a 10–4 diluted 10% brain homogenate was performed by the Kaplan-Meier method.25 Kaplan-Meier survival curves were plotted for the IL-1RI–/– and wild-type mice (data not shown). Differences between the survival curves were statistically evaluated using the log rank test.26 In addition, the survival in all groups was statistically analyzed using the unpaired t-test.
Northern Blot Analysis
RNA, isolated from scrapie-infected and mock-infected mice brains by acid guanidinium thiocyanate-phenol-chloroform extraction,27 was fractionated by agarose gel electrophoresis, transferred to nylon membranes, and hybridized with probes specific for IL-1ß, IL-1RI, IL-1ß-converting enzyme (ICE), glial fibrillary acidic protein (GFAP), and actin. The GFAP-specific probe was obtained from a cDNA library enriched for up-regulated genes in scrapie-infected brain tissue as previously described,6 all other DNAs used as hybridization probes were amplified by polymerase chain reaction (PCR) starting from murine spleen cDNA. Applied oligonucleotides for the PCR amplifications were: IL-1ß(+) CCTGTGTCTTTCCCGTG; IL-1ß(–) GAGTGCTGCCTAATGTCC; IL-1RI(+) GGGACTCCAGGATTCATCAG; IL-1RI(–) GCTCTTCAGCCACATTCC; ICE(+) CGTGGAGAGAAACAAGGAG; ICE(–) CAGCAGTGGGCATCTGTAG; actin(+) AAGGCCAACCGTGAAAAGAT; actin(–) CGCTCGTTGCCAATAGTGAT.
Quantitation of Chemokine mRNA Levels
RNAs were reverse-transcribed using the First-Strand cDNA synthesis kit (Amersham, Freiberg, Germany). Gene expression levels from three mice per group were subsequently determined by TaqMan real-time PCR using an GeneAmp 5700 sequence detection system (Perkin Elmer, Boston, MA) and the SYBR Green PCR kit (Qiagen, Hilden, Germany). Amplifications were performed with the primer sets mig-1 TTTTGGGCATCATCTTCCTG and mig-2 TTCCCCCTCTTTTGCTTTTT for CXCL9; and IP10-1 GCAACTGCATCCATATCGATGAC combined with IP10-2 TGTGCGTGGCTTCACTCCA for CXCL10, respectively. To ensure equivalence of the samples results were normalized for glyceraldehyde-3-phosphate dehydrogenase mRNA levels, which were detected using the primers GAPDH-1 GACCTCACCATCCCGCATCT and GAPDH-2 GCGGGAGTCGGCCAGTTACC.
Immunohistochemistry and PET Blot Analysis
All immunohistochemistry was performed according to previously published procedures.28 Examined areas were the cortex, hippocampus, brain stem, cerebellum, and the olfactory bulb from three mice per group and time point. Differences between IL-1RI–/– mice and the C57/B6 controls were evaluated by independent scoring of all tissue samples by three investigators without previous knowledge of the group to which the mice belonged. Antibodies used were a polyclonal anti-GFAP antiserum (DAKO, Glostrup, Denmark) and a monoclonal antibody against the CD 11b (Mac-1) antigen (kindly provided by B. Engelhardt, Munster, Germany). The prion protein was detected using the anti-PrP monoclonal antibody 6H4 (Prionics, Schlieren, Switzerland).
The PET blot analysis was performed as previously described.29 Briefly, 6-µm paraffin-sections of formalin-fixed brain tissues were transferred onto a nitrocellulose membrane and, after dewaxing, treated with proteinase K for digestion of normal PrPC. Remaining membrane-bound protein was subsequently denatured with 3 mol/L guanidine isothiocyanate. The actual immunodetection was performed with the anti-PrP 6H4 antibody followed by incubation with an alkaline phosphatase-linked anti-mouse immunoglobulin antiserum. The final staining was performed with nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate.
Results
As a first step we determined whether IL-1ß, ICE, which is required to generate active IL-1ß,30
and IL-1 receptor type 1 (IL-1RI) are expressed in a well-characterized murine scrapie model system; the scrapie-strain 139A-infected C57/B6 mouse.31
Northern blot analysis of total brain RNA showed the up-regulation of the respective mRNA levels in the scrapie-infected brain in comparison to brain tissue from age- and sex-matched mock-infected controls (Figure 1)
. The analyzed components of the IL-1 system were hardly expressed in the normal brain, whereas in the course of the infection the expression levels increased fourfold to sixfold.
|
and IL-1ß are mediated via binding to IL-1RI. The only other known IL-1 receptor, IL-1R type 2, lacks an intracellular domain necessary for signal transduction and is believed to act as a decoy receptor.32
To assess the role of IL-1 and IL-1ß during a prion disease of the central nervous system, the intracerebral scrapie infection of mice deficient for the IL-1RI33
was compared to similarly infected wild-type C57/B6 mice. The mean survival time for the scrapie-infected IL-1RI–/– mice was 197 (±7) days in comparison to 172 (±4) days for C57/B6 mice receiving the same infectious dose using a 10–2 diluted brain homogenate prepared from a terminally ill animal as inoculum (Table 1)
. Moreover, C57/B6 mice intracerebrally infected on the same day with a 10–3 dilution of the same brain homogenate survived for 181 (±13) days on average (Table 1)
. Hence, the significantly (P < 0.001) increased resistance of the IL-1RI–/– mice is equivalent to the survival of wild-type mice, which received a more than 10 times lower challenge dose. In separate experiments IL-1RI–/– and C57/B6 mice were intracerebrally infected with a 10–4-diluted brain homogenate (Table 1)
. Again, the incubation periods of the IL-1RI–/– [226 ± 14 days after infection (dpi)] were significantly longer (P < 0.001) than those of the wild-type controls (204 ± 7 dpi). Furthermore, the survival time spread increased from 26 days for the controls to 43 days for the IL-1RI–/– mice, which additionally indicates the higher resistance of the IL-1RI-deficient mice against scrapie pathogenesis.12
|
The next parameter investigated was the scrapie-associated astrocytosis in IL-1RI–/– compared to C57/B6 mice, which was examined by immunohistochemistry in the asymptomatic stage of the disease (125 dpi) and in terminally ill animals. At the terminal stage of the disease no significant difference concerning activated, GFAP-expressing, astrocytes was seen between wild-type controls and the IL-1RI–/– mice. However, at 125 dpi, a pronounced astrocytosis in the cerebellum, olfactory bulb, and the brain stem was readily detectable in C57/B6 mice, whereas in IL-1RI–/– mice GFAP-positive astrocytes were nearly absent at this asymptomatic stage of the infection (Figure 2)
. A further comparison of terminally diseased wild-type controls (169 dpi) with still hardly symptomatic IL-1RI–/– mice sacrificed at the same day confirmed these observations, but the difference in GFAP-positive astrocytes between the two groups was at this time point only twofold to fourfold in the areas examined (data not shown). Additional Northern blot analysis of total brain RNA demonstrated the presence of fourfold to sixfold elevated GFAP mRNA levels at 125 dpi in the scrapie-infected C57/B6 mice. In contrast, GFAP expression in brain tissue of IL-1RI–/– mice at the same stage of the infection was still comparable to the mock-infected controls (Figure 2)
. The astrocytosis was further investigated by quantitative reverse transcriptase-PCR expression analysis of the CXC chemokines CXCL9 and CXCL10, which are in the central nervous system predominantly produced by reactive astrocytes.34,35
CXCL9 and CXCL10 mRNA levels were found to be 4.2 and 7.4 times higher at 125 dpi in the wild-type control brains compared to the IL-1RI–/– mice, respectively. At the terminal stage of the disease the expression of CXCL9 and CXCL10 was 8.3- and 4.6-fold increased in the scrapie-infected C57/B6 controls (Figure 3)
.
|
|
.
|
, but both groups appeared primarily similar at the terminal stage of the disease (data not shown). The PET blot technique facilitates the depiction of larger brain areas and allows for a stringent discrimination between normal PrP and proteinase K-resistant PrPSc. In agreement with the immunohistochemistry, PET blots from animals of the IL-1RI–/– mice at 125 dpi showed significantly less PrPSc accumulation in the cortex and thalamus region than the wild-type controls. In the terminal stage, the PrPSc accumulations of the disease were primarily indistinguishable between the two groups and were particularly intense in the cortex and the hippocampus (Figure 5)
.
|
The data presented here show that essential components of the IL-1 system are activated in the brains of scrapie-infected mice (Figure 1)
. The overexpression of IL-1ß (Figure 1)
is in good agreement with previous observations in various murine scrapie models,5,13,20,21
but contrasts the apparent lack of IL-1ß production in scrapie strain ME7-infected C57/B6 mice.37
Interestingly, Campbell and colleagues20
described a predominant induction of IL-1 expression in the Chandler/SWRj scrapie model, whereas an increase of IL-1ß levels was evident but comparatively low, which may require very sensitive methods to be detectable.21
Thus, besides possible sensitivity issues, to look for expression levels of a single component of the IL-1 system does not necessarily inform about its functional activation.
The additional induction of ICE and IL-1RI is likely to contribute significantly to the biological activity of IL-1 as more proIL-1ß, is converted into its active form and increased receptor expression renders cells more responsive to IL-1 binding. The combined effects of this IL-1 system activation may well lead to a considerably higher degree of available IL-1 bioactivity than the Northern blot data in Figure 1
indicate. To identify the functions of IL-1 in a model system for chronic neurodegeneration we investigated the scrapie infection of IL-1RI–/– mice in comparison to C57/B6 controls.
The differences between the IL-1RI–/–-deficient mice and the controls concerning the scrapie-associated astrocyte activation were striking (Figure 2)
. At the early stage of the infection the astrocytosis in the C57/B6 group greatly exceeded the astroglia activation in the IL-1RI–/– mice. The data indicate that IL-1 is a driver of the early scrapie-associated astrocytosis in vivo. Given that in the late stage of the infection the staining of GFAP-positive reactive astrocytes was similar for wild-type controls and IL-1RI–/– mice, other factors, among them possibly IL-6, which is up-regulated during scrapieinfections5,13
and activates astrocytes in vitro,11
appear to compensate for the absence of IL-1 bioactivity in astrocyte activation. In agreement with this hypothesis, similar effects were observed for the activation of astrocytes throughout time in IL-1RI–/– mice after brain injury.15
Furthermore, it was recently shown that astrocytes interact directly with amyloid fibrils via scavenger receptors yet to be identified.38
Thus, the activation of astrocytes may also be promoted by scavenger receptor-mediated signaling as a consequence of the ongoing deposition of misfolded PrPSc. This activating stimulus could represent a further mechanism to compensate for the absence of IL-1 bioactivity in the IL-1RI–/– mice.
In addition to using GFAP detection to determine the extent of astrogliosis in the scrapie-infected mice we assessed the expression of the chemokines CXCL9 and CXCL10, known to be produced by activated astrocytes. As expected, the expression of CXCL9 and CXCL10 was significantly elevated in the wild-type controls compared to the IL-1RI–/– mice (Figure 3)
. However, even in the terminal stage of the disease, which showed no changes in GFAP expression, CXCL9 and CXCL10 mRNA levels were 8.3 and 4.6 times higher in the C57/B6 control group than in the IL-1RI–/– mice (Figure 3)
. This suggests that, in addition to the quantitative differences observed, the activation status of astrocytes (and hence the chemokine/cytokine profile) differs in the presence or absence of IL-1 bioactivity. It was previously shown that IL-1 is a strong inducer of CXCR3 ligand expression in astrocytes.39,40
The sustained increase of astrocytic CXCL9 and CXCL10 mRNA levels in the wild-type mice compared to the IL-1RI–/– group is therefore most likely because of the direct induction of chemokine expression by IL-1. Given that the chemokine receptor CXCR3 is expressed on microglia, astrocytes, and neurons it is well possible that CXCR3 ligands are involved in glial and glial/neuronal interactions.41
CXCR3 ligands may trigger the migration of microglia toward regions of neuronal damage,42
and could thereby promote the activation of glial cells by degenerating nervous tissue. Moreover, CXCL9 and CXCL10 activate the ERK1/2 pathway in murine neurons and sustained ERK1/2 activation was suggested to play a detrimental role in acute and chronic neurodegeneration.41
The infiltration of T cells, which are most likely attracted by CXCR3 ligands of astrocytic origin, into the brain in murine and human prion diseases is a further consequence of chemokine overexpression with a possible influence on the disease development.43
Several other aspects link the astrocytosis to scrapie pathogenesis. First of all, in vitro experiments demonstrated that astrocyte activation is associated with the secretion of neurotoxic factors.10
Thus, the scrapie-associated, in part IL-1-driven, astrocytosis could promote disease development. Furthermore, the observation that PrP-expression specifically targeted to astrocytes is sufficient to render resistant PrP knockout mice susceptible to scrapie suggests that PrP-expressing astrocytes are a relevant reservoir for prion replication in vivo.44
In addition, the accumulation of PrPSc was previously described to occur in astrocytes in scrapie-infected mice and hamsters.45,46
Therefor, the delayed PrPSc deposition in the IL-1RI–/– mice may be linked to the reduced astrocytosis in these animals (Figures 2, 3, and 5)
.
As outlined in the introduction, microglia are likely to be initially activated by the PrPSc deposition. Accordingly, both the IL-1RI–/– mice and the wild-type controls developed a vigorous microgliosis in response to the ongoing PrPSc accumulation (Figures 4 and 5)
. Thus, IL-1 is not essential for microglia activation during infections with the scrapie agent. In contrast to this model of an amyloid-driven microgliosis the microglia activation in acute neurodegeneration after brain injury is primarily IL-1-dependent.15
It is however less clear why the microglia activation is moderately more pronounced in the IL-1RI–/– mice than in the wild-type controls. Interestingly, microglia activation and function can be negatively affected by soluble factors produced by astrocytes as well as by direct astrocytic contact to microglia.47-50 Hence, the more pronounced microgliosis in the IL-1RI–/– mice could be because of the reduced astrocytosis in these animals and possibly affects the deposition of PrPSc by enhanced microglial phagocytosis of amyloid.51-54
Our observations indicate that IL-1 acts as a driver of the scrapie-associated activation of astrocytes during early stages of the disease. Although IL-1 is clearly not essential for the development of the ultimately lethal neurodegeneration (Table 1)
, the experimental model suggests that therapeutic agents that suppress astrocyte activation and/or glial IL-1 expression may help to delay disease onset in established prion infections of the central nervous system. In addition, our study encourages further research efforts to identify other host factors, which participate in disease development and in particular in the deleterious inflammatory aspects of the disease-associated gliosis.
Acknowledgements
We thank K. Krohn, S. Lichy, and H. Wohlert for expert technical assistance; U. Erikli for critical reading of the manuscript; and U. Mönning and N. Holtkamp for helpful discussions.
Footnotes
Address reprint requests to Dr. Michael Baier, Project "Neurodegenerative Diseases," Robert-Koch-Institute, Nordufer 20, 13353 Berlin, Germany. E-mail: baierm{at}rki.de
Supported in part by the Federal Ministry for Health and Social Security (grant 325-4471-02/45) and the Federal Ministry for Education and Research, Germany (grant 01KO0111).
Accepted for publication April 12, 2004.
References
This article has been cited by other articles:
 |
|
 |
|
  C. Riemer, J. Schultz, M. Burwinkel, A. Schwarz, S. W. F. Mok, S. Gultner, T. Bamme, S. Norley, F. van Landeghem, B. Lu, et al. Accelerated Prion Replication in, but Prolonged Survival Times of, Prion-Infected CXCR3-/- Mice J. Virol., December 15, 2008; 82(24): 12464 - 12471. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. O'Shea, E. G. Maytham, J. M. Linehan, S. Brandner, J. Collinge, and S. E. Lloyd Investigation of Mcp1 as a Quantitative Trait Gene for Prion Disease Incubation Time in Mouse Genetics, September 1, 2008; 180(1): 559 - 566. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Tamguney, K. Giles, D. V. Glidden, P. Lessard, H. Wille, P. Tremblay, D. F. Groth, F. Yehiely, C. Korth, R. C. Moore, et al. Genes contributing to prion pathogenesis J. Gen. Virol., July 1, 2008; 89(7): 1777 - 1788. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Riemer, M. Burwinkel, A. Schwarz, S. Gultner, S. W. F. Mok, I. Heise, N. Holtkamp, and M. Baier Evaluation of drugs for treatment of prion infections of the central nervous system J. Gen. Virol., February 1, 2008; 89(2): 594 - 597. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. S. Rambold, M. Miesbauer, D. Rapaport, T. Bartke, M. Baier, K. F. Winklhofer, and J. Tatzelt Association of Bcl-2 with Misfolded Prion Protein Is Linked to the Toxic Potential of Cytosolic PrP Mol. Biol. Cell, August 1, 2006; 17(8): 3356 - 3368. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Vet. Pathol., November 1, 2004; 41(6): 712 - 712. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
