- Email: [email protected]
Activation of nuclear factor-κB by β-amyloid peptides and interferon-γ in murine microglia
Journal of Neuroimmunology 77 Ž1997. 51–56
Activation of nuclear factor-k B by b-amyloid peptides and interferon-g in murine microglia Corrada Bonaiuto
1, 2
, Patrick P. McDonald 2 , Filippo Rossi, Marco A. Cassatella
)
Department of General Pathology, UniÕersity of Verona, Strada Le Grazie 4, 37134 Verona, Italy Received 7 January 1997; accepted 7 February 1997
Abstract An increasing body of evidence suggests that amyloid-b ŽA b . peptides and microglia are crucially involved in the pathogenesis of Alzheimer’s disease. In an effort to further elucidate the biological effects of A b towards microglia, we investigated the ability of A b peptides to activate nuclear factor ŽNF.-k B in the N9 murine microglial cell line. Co-stimulation of microglia with suboptimal concentrations of A b Ž25–35. and 100 Urml IFNg resulted in the detection of a specific NF-k B DNA-binding activity in nuclear extracts, as determined in gel mobility shift assays. This response required at least 120 min to be evident and supershift experiments revealed that the NF-k B complex contains both RelA and p50. Accordingly, immunoblot experiments showed that amongst NF-k BrRel proteins, RelA and p50 are mobilized to the nucleus following microglial cell stimulation with A b Ž25–35. plus IFNg . Higher concentrations of A b Ž25–35. were effective by themselves in inducing NF-k B activation, both in the N9 microglial cell line and in rat primary microglia, as well as in human monocytes. For purposes of comparison, microglia were also stimulated with bacterial LPS, a known NF-k B inducer. As expected, LPS strongly induced the formation of two NF-k B DNA-binding activities, one of which was identified as RelArp50. The LPS response was also more rapid, as it was already evident by 40 min and remained sustained for up to 3 h. Collectively, these findings indicate that NF-k B activation might constitute one of the mechanisms underlying the inducible expression of k B-dependent genes in microglia stimulated by A b peptides and IFNg , or by LPS. Keywords: Amyloid-b ; NF-k B; Transcription factors; Microglia ; Alzheimer’s disease
1. Introduction Amyloid-b peptide ŽA b . is a 39- to 42-amino acid cleavage product of a transmembrane glycoprotein, amyloid precursor protein ŽAPP.. Abundant extracellular deposits of A b are typically observed in various regions of the cerebral cortex and in the hippocampus of patients afflicted with Alzheimer’s disease ŽAD. and to a lesser extent, in normally aging individuals as well ŽMullan and Crawford, 1993; Selkoe, 1994.. These A b-rich aggregates Žalso known as senile plaques. are characteristically infiltrated and surrounded by activated microglia and astrocytes and are usually associated with local cortical atrophy and extensive neuronal loss ŽSelkoe, 1994.. Whereas the
) Corresponding author. Tel.: q39-45-8098130; fax: q39-45-8098127; e-mail: [email protected] 1 Current affiliation: Istituto di Patologia Generale, Universita` di Catania, Italy. 2 Acknowledgement of the equal contribution of the first two authors.
pathogenesis of AD remains poorly understood, the above considerations indicate a potentially important role for A b peptide in the neurodegenerative process. In keeping with this view, both full-length A b and a fragment spanning amino acids 25–35 of the molecule ŽA b Ž25–35.. have been reported to exert a direct cytotoxic effect on neurons in primary hippocampal cultures ŽYankner et al., 1990; Pike et al., 1991, 1993. and to induce neuronal degeneration in adult rat brain in vivo ŽKowall et al., 1991.. In addition to these direct effects, A b can also indirectly promote neuronal damage by virtue of its ability to activate microglia. In this respect, recent studies from our laboratory have demonstrated that co-stimulation of murine microglial cells or human monocytes with A b and IFNg results in the generation of reactive nitrogen intermediates, as well as in the production of pro-inflammatory cytokines such as TNFa , IL-8 and monocyte chemotactic protein-1 ŽMCP-1.rJE ŽMeda et al., 1995, 1996.. Moreover, co-culture experiments established that activation of microglia with A b and IFNg led to neuronal cell injury ŽMeda et al., 1995..
0165-5728r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 5 7 2 8 Ž 9 7 . 0 0 0 5 4 - 4
52
C. Bonaiuto et al.r Journal of Neuroimmunology 77 (1997) 51–56
Despite an increasing body of evidence in support of a role for A b and microglia in neurodegenerative processes, the molecular mechanisms underlying the action of A b towards microglia remain to be elucidated. In this regard, it is noteworthy that the gene encoding the key enzyme involved in nitric oxide generation, namely inducible nitric oxide synthase ŽiNOS., as well as those encoding TNFa and MCP-1rJE, all feature k B or k B-like enhancer sequenceŽs. in their promoter region, which are important for inducible gene expression ŽCollart et al., 1990; Shakhov et al., 1990; Lowenstein et al., 1993; Ueda et al., 1994.. Increased transcriptional activity results from the binding to these enhancer sequences of nuclear factor ŽNF.-k B, an ubiquitous transcription factor normally sequestered in the cytoplasm, but which can be mobilized to the nucleus following appropriate cell stimulation Žfor a review, see Bauerle and Henkel, 1994.. In the present study, we ¨ investigated whether A b peptides have the ability to activate the NF-k B pathway in murine microglia. We now report that A b has the ability to induce NF-k B activation and that co-stimulation of microglial cells or monocytes with IFNg results in an enhancement of this response. This therefore indicates that mobilization of the NF-k B pathway may underlie the action of A b towards cytokine production and inducible nitric oxide generation in microglial cells ŽMeda et al., 1995; Goodwin et al., 1995..
2. Materials and methods 2.1. Reagents and antisera The A b Ž25–35. peptide ŽGSNKGAIIGLM., as well as the full-length A b Ž1–42. peptide, were kindly provided by Dr. L. Otvos ŽWistar Institute, Philadelphia, PA. ŽOtvos et al., 1993.; the peptides were dissolved in sterile PBS, aliquoted and kept at y208C. Recombinant murine IFNg was obtained from Dr. G. Garotta ŽHoffmann-LaRoche, Basel.. Rabbit antisera to human c-Rel Ža265 and a1136, raised against the C-terminal region and against an internal sequence downstream from the nuclear localization signal, respectively., p65rRelA Ža1207 and a1226, against the N-terminal and C-terminal regions, respectively. and p50rNFk B1 Ža1141, against the N-terminal region., were a generous gift from Dr. N.R. Rice ŽNCI-Frederick Cancer Research and Development Center, Frederick, MD.. The specificity of these antisera has already been extensively characterized ŽRice et al., 1992; Rice and Ernst, 1993.. A commercial antibody directed against the nuclear localization sequence of p50 Ža114X. was also purchased from Santa Cruz Biotechnology ŽSanta Cruz, CA.. An oligonucleotide containing tandemly repeated NF-k B sites identical to those of the HIV promoter Ž5X-GATCAGGGACTTTCCGCTGGGGACTTTCC-3X ; underlined is the NF-k B binding sequence. was kindly provided by Dr. G. Trinchieri ŽWistar Institute, Philadelphia, PA.. Poly ŽdI-dC., T4
polynucleotide kinase and Sephadex G25 spin columns were purchased from Pharmacia ŽUppsala, Sweden.. Horseradish peroxidase-linked donkey anti-rabbit antibody, the enhanced chemoluminescence ŽECL. detection kit and w g-32 Px-ATP were purchased from Amersham ŽLittle Chalfont, England.. RPMI 1640 was from GIBCOrBRL ŽGaithersburg, MD. and low-endotoxin FCS Ž- 6 pgrml. from Hyclone ŽLogan, Utah.. Polystyrene flasks and plates for cell culture were from NUNC ŽRotskilde, Denmark.. Aprotinin, acetylated BSA, leupeptin, lipopolysaccharide ŽLPS, from E. coli strain 026B6., Nonidet P40 ŽNP40., pepstatin A and phenylmethanesulphonyl fluoride ŽPMSF. were all from the Sigma Chemical Co. ŽSt. Louis, MO.. All other reagents were molecular biology grade and all buffers and solutions were prepared using pyrogen-free clinical grade water. 2.2. Cell culture The N9 murine microglial cell line was generously provided by Dr. P. Ricciardi-Castagnoli ŽCentro di Citofarmacologia, C.N.R., Milan. ŽRighi et al., 1989.. Microglial cells were cultured in RPMI 1640 supplemented with 10% heat-inactivated FCS, 2 mM L-glutamine, 50 Urml penicillin and 50 m grml streptomycin, at 378C under a 5% CO 2 atmosphere, at a final concentration of 10 6 cellsrml. Rat primary microglia were kindly provided by Dr. U. Armato ŽDepartment of Human Anatomy, University of Verona. and were isolated from Wistar rats and cultured as described previously for murine microglial cells ŽChao et al., 1992.. The rat microglia used in this study were ) 90% pure, as determined by immunocytochemistry. Human peripheral blood monocytes were purified and cultured as described previously ŽMeda et al., 1996.. Cultured cells were exposed to various concentrations of A b Ž25– 35., 50 m grml of ‘aged’ A b Ž1–42. ŽMeda et al., 1995; Pike et al., 1991., or 100 ngrml LPS Žor their diluent, RPMI 1640., for the indicated times. 2.3. Nuclear extract preparation and electrophoretic mobility shift assays After the desired incubation time with the stimuli, microglial cells were washed twice with ice-cold PBS and then harvested using a rubber policeman in 1 ml of lysis buffer Ž10 mM Hepes pH 7.9, 10 mM KCl, 0.2 mM EDTA, 0.2 mM EGTA, 1 mM DTT. supplemented with protease inhibitors Ž0.5 mM PMSF and 10 m grml each of aprotinin, leupeptin and pepstatin A.. After a 15 min incubation on ice, NP40 was added to a final concentration of 0.1% Žvrv. and the samples were vigorously vortexed, prior to centrifugation Ž12,000 g, 1 min, 48C.. Nuclear pellets were resuspended in 30 m l of ice-cold nuclear extraction buffer Ž20 mM Hepes pH 7.9, 400 mM NaCl, 0.2 mM EDTA, 0.2 mM EGTA, 1 mM DTT. containing the antiprotease cocktail and kept under vigorous agitation
C. Bonaiuto et al.r Journal of Neuroimmunology 77 (1997) 51–56
for 15 min at 48C. Samples thus treated were centrifuged Ž12,000 g, 10 min, 48C. and nuclear extracts were aliquoted and stored at y808C. Small aliquots of the extracts were routinely processed for protein content determination. Electrophoretic mobility shift assays ŽEMSA. were performed essentially as described earlier ŽMcDonald et al., 1995.; samples Žcontaining 5 m g of nuclear extract. were electrophoresed on 5% polyacrylamide gels Ž2 h, 200 V, 48C. in 0.5 = TBE and dried gels were exposed to Kodak X-O-Mat film with intensifying screens for periods varying between 6 and 18 h. 2.4. Immunoblots Nuclear extracts were dissolved 2-fold by the addition of an equal volume of sample buffer 2 = Ž50 mM TrisBase pH 6.80, 4% sodium dodecylsulfate Žwrv., 10% 2-mercaptoethanol Žvrv., 20% glycerol Žvrv.., heated for 5 min at 958C and used immediately or stored at y208C. Samples were electrophoresed on 15% Kornberg gels ŽThomas and Kornberg, 1975.. Proteins were then transferred onto a nitrocellulose membrane at 20 V constant for 60 min in a Transblot semidry transfer cell ŽBioRad.. Transfer efficiency was visualized by reversible Ponceau Red staining. The membranes were then processed essentially as described earlier ŽPouliot et al., 1994.. Primary antibodies were used at the following dilutions: anti-RelA Ža1207. or anti-p50 Ža1141., 1:2000; anti c-Rel Ža265., 1:20,000. 3. Results We first examined the ability of A b and of other stimuli to induce NF-k B DNA-binding activity in the
53
murine microglial N9 cell line. For this purpose, cells were cultured in the presence or absence of A b Ž25–35., IFNg , LPS, or combinations thereof, for varying lengths of time. Nuclear proteins were extracted and the NF-k B DNA-binding activity of the extracts was assessed in EMSA. As shown in Fig. 1A, stimulation of microglia with either 100 Urml IFNg or 20 m grml A b Ž25–35. exerted little effect towards NF-k B DNA-binding activity, up to 3 h Žlanes 1, 5, 9 and 2, 6, 10, respectively.. However, co-stimulation of the cells with A b Ž25–35. and IFNg led to a substantial increase in nuclear NF-k B DNA-binding activity, which consisted of a single prominent complex Žband ‘A’. and required at least 120 min to become evident ŽFig. 1A, lane 11.. Similar results were obtained using a commercial A b Ž25–35. peptide ŽSigma., in which the C-terminus was not amidated Ždata not shown.. By contrast, LPS strongly induced the formation of two distinct NF-k B complexes Žbands ‘A’ and ‘B’.; this effect was rapid, as it was already evident after 40 min of stimulation and was sustained for up to 3 h ŽFig. 1A, lanes 4 and 12.. Similar results were obtained in cells stimulated with 100 Urml TNFa Ždata not shown.. In a separate series of experiments, we examined the effect of higher concentrations of A b Ž25–35. towards NF-k B activation in microglial N9 cells. For purposes of comparison, N9 cells were also stimulated with equivalent concentrations of full-length A b Ž1–42., both in the presence and absence of IFNg . Fig. 1B shows that at a concentration of 50 m grml, A b Ž25–35. was able to directly induce NF-k B DNA-binding activity Žlane 2., whereas A b Ž1–42. exerted but a minor effect. Under these conditions, co-stimulation of N9 cells with IFNg markedly enhanced the effect of both A b peptides towards NF-k B activation ŽFig. 1B, lanes 3 and 5..
Fig. 1. ŽA. Time course of NF-k B DNA-binding activity induction by A b Ž25–35., IFNg and LPS in the N9 microglial cell line. Cells were cultured for the indicated times in the presence or absence of 20 m grml A b Ž25–35., 100 Urml IFNg , 100 ngrml LPS, or combinations thereof. Nuclear extracts were then prepared and analyzed in EMSA, as described in Section 2. The amount of nuclear extract used in the binding reactions corresponded to 5 m g of protein. This experiment is representative of four: , unstimulated microglia; A b , A b Ž25–35.-stimulated cells; IFN, IFNg-stimulated cells. ŽB. Comparative effect of A b Ž25–35. and A b Ž1–42. towards nuclear NF-k B DNA-binding activity in IFNg-treated or diluent-treated N9 microglial cells. Cells were cultured for 3 h with 50 m grml A b Ž25–35. or 50 m grml A b Ž1–42., in the presence or absence of 2 ngrml IFNg , prior to EMSA analysis of the resulting nuclear extracts. ctrl, unstimulated microglia; 25–35, A b Ž25–35.-stimulated cells; 1–42, A b Ž1–42.-stimulated cells.
54
C. Bonaiuto et al.r Journal of Neuroimmunology 77 (1997) 51–56
Fig. 2. Characterization of the NF-k B DNA-binding activities present in nuclear extracts of activated N9 microglial cells. ŽA. Nuclear extracts Ž5 m g. from microglial cells stimulated for 3 h with 20 m grml A b Ž25–35. and 2 ngrml IFNg were incubated in the absence Ž‘y’. or in the presence of a 25-fold molar excess of unlabeled oligonucleotide probe Ž‘cold’., prior to the addition of the labeled NF-k B probe to the binding mixtures and subsequent EMSA analysis. ŽB. The same nuclear extracts were incubated in the absence Ž‘y’. or in the presence of specific antisera raised against c-Rel Ža1136., p65rRelA Ža1226., or p50 Ža114X., prior to the addition of the labeled NF-k B probe and subsequent EMSA analysis. As a control, nuclear extracts were also incubated in the presence of pre-immune rabbit serum Ž‘NRS’.. The experiments depicted in this figure are representative of three ŽA. and two ŽB..
We next sought to determine the specificity of the inducible NF-k B complexes detected in microglial cell nuclear extracts. As shown in Fig. 2A, these inducible NF-k B complexes showed specificity in their interaction with our NF-k B oligonucleotide probe, as they were no longer detectable when a 25-fold molar excess of unlabeled probe was included in the binding mixture Žlane 2.; similar results were obtained using nuclear extracts from LPS-treated microglia Ždata not shown.. We also investigated the subunit composition of the inducible NF-k B complexes present in nuclear extracts of activated N9 microglia. To this end, supershift experiments were performed, using specific antisera raised against individual NF-k BrRel proteins. As shown in Fig. 2B, antisera to either RelA or p50 efficiently supershifted the major inducible complex present in extracts from IFNg- plus A b Ž25–35.-treated microglia Žlanes 2 and 3.. In LPStreated cells, the NF-k B DNA-binding activity co-migrating with the aforementioned complex Žband ‘A’. was similarly affected by anti-RelA and anti-p50 antisera Ždata not shown.. By contrast, an antiserum to c-Rel consistently failed to affect the detection or mobility of this NF-k B DNA-binding activity ŽFig. 2B, lane 4.. Our inability to detect c-Rel-containing complexes in supershift experiments prompted us to examine which NF-k BrRel proteins are present in nuclear extracts from activated microglia. Fig. 3 shows that nuclear fractions from unstimulated microglia contain little, if any, immunoreactive p50 or RelA Žlane 1.. Following a 3 h exposure to either IFNg or
A b Ž25–35. as single stimuli, little or no detectable changes in the nuclear levels of p50 or RelA were observed ŽFig. 3, lanes 2 and 3.. However, co-stimulation of microglia with IFNg and A b Ž25–35. consistently resulted in the mobilization of both proteins to the nuclear compartment ŽFig. 3, lane 4.. By comparison, stimulation of microglia for 3 h with LPS led to a stronger nuclear accumulation of both p50 and RelA ŽFig. 3, lane 5.. Under all conditions tested, nuclear c-Rel was undetectable; this could not be attributed to a failure of our c-Rel antiserum to recognize the protein, since immunoreactive c-Rel was readily detected in cytoplasmic extracts Ždata not shown.. In a final series of experiments, we investigated whether the data obtained in the N9 microglial cell line could be reproduced using primary microglial cells. For this purpose, primary rat microglia were isolated and cultured for 3 h in the presence or absence of either 50 m grml A b Ž25–35., 100 Urml IFNg , or both. Fig. 4A shows that by itself, A b Ž25–35. has the ability to substantially induce NF-k B activation Žlane 3. and that co-stimulation of the cells with IFNg resulted in a further Žalbeit modest. enhancement of this response Žlane 4.. Similarly, we investigated whether A b-elicited NF-k B activation might also be observed in cells of human origin. For this purpose, we used human peripheral blood monocytes, since we have previously shown that they respond to A b Ž25–35. in a similar manner to microglia in many respects ŽMeda et al., 1996.. As shown in Fig. 4B, a 3 h culture of human monocytes in the presence of 40 m grml A b Ž25–35. resulted in the induction of a prominent NF-k B DNA-binding activity Žlane 2. and co-stimulation of the cells with IFNg somewhat increased the extent of this response Žlane 4.. Thus, the ability of A b peptides to induce NF-k B activation is not singular feature of cell lines Žsuch as N9
Fig. 3. Nuclear mobilization of NF-k BrRel proteins in response to A b Ž25–35., IFNg and LPS in the N9 microglial cell line. Cells were cultured for 3 h in the presence or absence of 20 m grml A b Ž25–35., 2 ngrml IFNg , 100 ngrml LPS, or combinations thereof. Nuclear extracts were then prepared and analyzed by immunoblotting using antisera to p65rRelA Ža1207. or p50 Ža1141., as described in Section 2. This experiment is representative of at least two. , unstimulated microglia; A b , A b Ž25–35.-stimulated cells; IFN, IFNg-stimulated cells; LPS, LPS-stimulated cells.
C. Bonaiuto et al.r Journal of Neuroimmunology 77 (1997) 51–56
Fig. 4. Activation of NF-k B by A b Ž25–35. and IFNg in rat primary microglia and primary human monocytes. ŽA. Rat primary microglia were cultured for 3 h in the presence or absence of 50 m grml A b Ž25–35. andror 2 ngrml IFNg . Nuclear extracts were then prepared and analyzed in EMSA, as described in Section 2. The amount of nuclear extract used in the binding reactions corresponded to 5 m g of protein. ŽB. Human monocytes were cultured for 3 h in the presence or absence of 40 m grml A b Ž25–35. andror 100 Urml IFNg , prior to EMSA analysis of the resulting nuclear extracts. , unstimulated cells; A b , A b Ž25– 35.-stimulated cells; IFN, IFNg-stimulated cells.
murine microglia., since it was also found to occur in primary rat microglia or primary human monocytes.
4. Discussion In recent years, it has become increasingly evident that both A b and microglial cells play a fundamental role in the pathogenesis of AD. In an effort to elucidate the biological effects of A b towards microglia, we have previously shown that a marked increase in the expression of the transcripts encoding TNFa and MCP-1rJE can be induced in murine N9 microglial cells or in human monocytes by A b Ž25–35. andror IFNg , resulting in the release of the corresponding proteins ŽMeda et al., 1995, 1996.. In view of the fact that NF-k B activation is crucially involved in the inducible expression of the genes encoding TNFa and MCP-1rJE ŽCollart et al., 1990; Shakhov et al., 1990; Ueda et al., 1994., we investigated whether the reported effects of A b Ž25–35. towards microglial gene expression might be paralleled by the activation of NF-k B. We now report that A b Ž25–35. has the ability to activate the NF-k B pathway in murine microglial cells, as well as in primary rat microglia and human monocytes. Although this property of A b Ž25–35. was particularly evident when it was used at higher concentrations, suboptimal concentrations were nevertheless found to synergize with IFNg in inducing NF-k B activation. Immunoblot experiments
55
showed that amongst NF-k BrRel proteins, RelA and p50 are mobilized to the nucleus following microglial cell stimulation with A b Ž25–35. plus IFNg . Accordingly, supershift experiments revealed that the inducible NF-k B DNA-binding activity observed under these conditions consisted of a major complex containing both p50 and RelA. Throughout our study, we also used the well characterized NF-k B activators, LPS and TNFa , for purposes of comparison. As expected, both stimuli were strong inducers of nuclear NF-k B DNA-binding activity, which consisted of at least two specific complexes, one of which being identified as p50rRelA heterodimers. Stimulation of the NF-k B pathway by LPS or TNFa Ži.e. NF-k B activation and the concurrent nuclear mobilization of RelA and p50. was also found to be more rapid and to occur on a greater scale, than that observed in response to A b Ž25–35. even when microglial cells were contemporaneously exposed to A b peptides and IFNg . Taken together, the qualitative and quantitative differences observed between the effects of LPS and those of A b peptides are likely to reflect the existence of distinct signal transduction pathways leading to NF-k B activation. On the basis of the present findings, it is tempting to speculate that NF-k B activation by A b peptides and IFNg might represent one of the molecular mechanisms underlying the inducible expression of the genes encoding TNFa , MCP-1rJE in murine microglia and human monocytes. Similarly, the APP promoter is known to contain two identical k B motifs ŽGrilli et al., 1995. and A b Ž25–35. was recently shown to elicit the transcriptional activation of a k B-driven reporter gene construct in a rat neuronal cell line ŽBehl et al., 1994.. The latter observations therefore raise the possibility, that conditions leading to NF-k B activation in murine microglia might also result in an enhanced expression of the APP gene and of the corresponding protein. In this respect, it is noteworthy that microglia not only constitutively release A b , but can also be induced to secrete substantial quantities of the peptide following stimulation with either LPS or A b Ž25–35. itself ŽBitting et al., 1996.. In a broader context, microglia have been proposed to represent an important source of A b in AD and in view of the above considerations, the combined presence of A b and microglia in senile plaques might contribute to maintain these cells in an activated state. In this context, the ability of A b Ž25–35. to induce the production of TNFa in microglia could lead to a persistent state of inflammation at sites of b-amyloid deposition, and it can be envisaged that the production of MCP-1rJE by activated microglia would further amplify this process through the recruitment of additional microglia to senile plaques. On a final note, we observed that in both murine and rat microglia, A b-elicited NF-k B activation could be markedly enhanced in the presence of IFNg , a cytokine that is primarily derived from activated T-cells and this was especially evident when microglia were exposed to suboptimal concentrations of A b peptides. Although T-
56
C. Bonaiuto et al.r Journal of Neuroimmunology 77 (1997) 51–56
lymphocytes are rarely observed in the central nervous system, they nevertheless have been reported to be present in the vicinity of senile plaques in AD patients, along with microglia ŽRogers et al., 1988; McGeer et al., 1991.. Thus, it is conceivable that microglial cell may activate NF-k B in response to relatively low quantities of A b peptides, through microglialrT-cell interactions. Collectively, the current findings suggest that therapeutical strategies aimed to control the development of AD could include the administration of drugs that hinder NF-k B activation.
Acknowledgements This work was supported by grants to MAC from the Consiglio Nazionale delle Ricerche Žprogetto strategico citochine 95.02809.ST74., from the M.U.R.S.T. Ž40% and 60%.. PPMcD is a post-doctoral Fellow of the Medical Research Council of Canada. We are especially grateful to Dr. Nancy Rice for having provided many antisera; we also wish to thank Dr. Ubaldo Armato for providing rat primary microglia and to Anette Bald, Ercolina Bianchini and Ilaria dal Pra for excellent technical assistance.
References Bauerle, P.A., Henkel, T., 1994. Function and activation of NF-k B in the ¨ immune system. Ann. Rev. Immunol. 12, 141–179. Behl, C., Davis, J.B., Lesley, R., Schubert, D., 1994. Hydrogen peroxide mediates amyloid-b protein toxicity. Cell 77, 817–827. Bitting, L., Naidu, A., Cordell, B., Murphy, G.M., 1996. b-Amyloid peptide secretion by a microglial cell line is induced by b-amyloid Ž25–35. and lipopolysaccharide. J. Biol. Chem. 271, 16084–16089. Chao, C.C., Hu, S., Molitor, T.W., Shaskan, E.G., Peterson, P.K., 1992. Activated microglia mediate neuronal cell injury via a nitric oxide mechanism. J. Immunol. 149, 2736–2741. Collart, M.A., Baeuerle, P., Vassalli, P., 1990. Regulation of tumor necrosis factor-a transcription in macrophages: Involvement of four k B-like motifs and of constitutive and inducible forms of NF-k B. Mol. Cell. Biol. 10, 1498–1506. Goodwin, J.L., Uemura, E., Cunnick, J.E., 1995. Microglial release of nitric oxide by the synergistic action of b-amyloid and IFNg . Brain Res. 692, 207–214. Grilli, M., Ribola, M., Alberici, A., Valerio, A., Memo, M., Spano, P., 1995. Identification and characterization of a k BrRel binding site in the regulatory region of the amyloid precursor protein gene. J. Biol. Chem. 270, 26774–26777. Kowall, N.W., Flint Beal, M., Busciglio, J., Duffy, L.K., Yankner, B.A., 1991. An in vivo model for the neurodegenerative effects of b amyloid and protection by substance P. Proc. Natl. Acad. Sci. USA 88, 7247–7251. Lowenstein, C.J., Alley, E.W., Raval, P., Snowman, A.M., Snyder, S.H.,
Russell, S.W., Murphy, W.J., 1993. Macrophage nitric oxide synthase gene: Two upstream regions mediate induction by interferon-g and lipopolysaccharide. Proc. Natl. Acad. Sci. USA 90, 9730–9734. McDonald, P.P., Cassatella, M.A., Bald, A., Maggi, E., Romagnani, S., Gruss, H.J., Pizzolo, G., 1995. CD30 ligation induces nuclear factork B activation in human T-cell lines. Eur. J. Immunol. 25, 2870–2876. McGeer, P., McGeer, E., Kawamata, H., Yamada, T., Akiyama, H., 1991. Reactions of the immune system in chronic degenerative neurological diseases. Can. J. Neurol. Sci. 18, 376–379. Meda, L., Cassatella, M.A., Szendrei, G.I., Otvos, L., Baron, P., Villalba, M., Ferrari, D., Rossi, F., 1995. Activation of microglial cells by b-amyloid protein and interferon-g . Nature 374, 647–650. Meda, L., Bernasconi, S., Bonaiuto, C., Sozzani, S., Zhou, D., Otvos, L., Mantovani, A., Rossi, F., Cassatella, M.A., 1996. b-Amyloid Ž25–35. peptide and IFN-g synergistically induce the production of the chemotactic cytokine, MCP-1rJE, in monocytes and microglial cells. J. Immunol. 157, 1213–1218. Mullan, M., Crawford, F., 1993. Genetic and molecular advances in Alzheimer’s disease. Trends Neurosci. 16, 398–403. Otvos, L., Szendrei, G.L., Lee, M.V., Mantsch, H.H., 1993. Human and rodent Alzheimer b-amyloid peptides acquire distinct conformations in membrane-mimicking solvents. Eur. J. Biochem. 211, 249–257. Pike, C.J., Walencewicz, A.J., Glabe, C.G., Cotman, C.W., 1991. In vitro aging of b-amyloid protein causes peptide aggregation and neurotoxicity. Brain Res. 563, 311–314. Pike, C.J., Burdick, D., Walencewicz, A.J., Glabe, C.G., Cotman, C.W., 1993. Neurodegeneration induced by b-amyloid peptides in vitro: The role of peptide assembly state. J. Neurosci. 13, 1676–1687. Pouliot, M., McDonald, P.P., Borgeat, P., McColl, S.R., 1994. Granulocyte-macrophage colony-stimulating factor stimulates the expression of the 5-lipoxygenase-activating protein ŽFLAP. in human neutrophils. J. Exp. Med. 179, 1225–1232. Rice, N.R., MacKichan, M.L., Israel, ` A., 1992. The precursor of NF-k B p50 has I k B-like functions. Cell 71, 243–253. Rice, N.R., Ernst, M.K., 1993. In vivo control of NF-k B activation by I k B-a . EMBO J. 12, 4685–4695. Righi, M., Mori, L., De Libero, G., Sironi, M., Biondi, A., Mantovani, A., Donini, S.D., Ricciardi-Castagnoli, P., 1989. Monokine production by microglial cell lines. Eur. J. Immunol. 19, 1443–1448. Rogers, J., Luber Narod, J., Styren, S.D., Civin, W.H., 1988. Expression of immune system-associated antigens by cells of the human central nervous system: Relationship to the pathology of Alzheimer’s disease. Neurobiol. Aging 9, 339–349. Selkoe, D.J., 1994. Cell biology of the amyloid-b protein precursor and the mechanism of Alzheimer’s disease. Annu. Rev. Cell. Biol. 10, 373–403. Shakhov, A.N., Collart, M.A., Vassalli, P., Nedospasov, S.A., Jongeneel, C.V., 1990. k B-type enhancers are involved in lipopolysaccharidemediated transcriptional activation of the tumor necrosis factor-a gene in primary macrophages. J. Exp. Med. 171, 35–47. Thomas, J.O., Kornberg, R.D., 1975. An octamer of histones in chromatin and free in solution. Proc. Natl. Acad. Sci. USA 72, 2626. Ueda, A., Okuda, K., Ohno, S., Shirai, A., Igarashi, T., Matsunaga, K., Fukushima, J., Kawamoto, S., Ishigatsubo, Y., Okubo, T., 1994. NF-k B and Sp1 regulate transcription of the human monocyte chemoattractant protein-1 gene. J. Immunol. 153, 2052–2063. Yankner, B.A., Duffy, L.K., Kirschner, D.A., 1990. Neurotrophic and neurotoxic effects of amyloid-b protein: Reversal by tachykinin neuropeptides. Science 250, 279–282.
Activation of nuclear factor-k B by b-amyloid peptides and interferon-g in murine microglia Corrada Bonaiuto
1, 2
, Patrick P. McDonald 2 , Filippo Rossi, Marco A. Cassatella
)
Department of General Pathology, UniÕersity of Verona, Strada Le Grazie 4, 37134 Verona, Italy Received 7 January 1997; accepted 7 February 1997
Abstract An increasing body of evidence suggests that amyloid-b ŽA b . peptides and microglia are crucially involved in the pathogenesis of Alzheimer’s disease. In an effort to further elucidate the biological effects of A b towards microglia, we investigated the ability of A b peptides to activate nuclear factor ŽNF.-k B in the N9 murine microglial cell line. Co-stimulation of microglia with suboptimal concentrations of A b Ž25–35. and 100 Urml IFNg resulted in the detection of a specific NF-k B DNA-binding activity in nuclear extracts, as determined in gel mobility shift assays. This response required at least 120 min to be evident and supershift experiments revealed that the NF-k B complex contains both RelA and p50. Accordingly, immunoblot experiments showed that amongst NF-k BrRel proteins, RelA and p50 are mobilized to the nucleus following microglial cell stimulation with A b Ž25–35. plus IFNg . Higher concentrations of A b Ž25–35. were effective by themselves in inducing NF-k B activation, both in the N9 microglial cell line and in rat primary microglia, as well as in human monocytes. For purposes of comparison, microglia were also stimulated with bacterial LPS, a known NF-k B inducer. As expected, LPS strongly induced the formation of two NF-k B DNA-binding activities, one of which was identified as RelArp50. The LPS response was also more rapid, as it was already evident by 40 min and remained sustained for up to 3 h. Collectively, these findings indicate that NF-k B activation might constitute one of the mechanisms underlying the inducible expression of k B-dependent genes in microglia stimulated by A b peptides and IFNg , or by LPS. Keywords: Amyloid-b ; NF-k B; Transcription factors; Microglia ; Alzheimer’s disease
1. Introduction Amyloid-b peptide ŽA b . is a 39- to 42-amino acid cleavage product of a transmembrane glycoprotein, amyloid precursor protein ŽAPP.. Abundant extracellular deposits of A b are typically observed in various regions of the cerebral cortex and in the hippocampus of patients afflicted with Alzheimer’s disease ŽAD. and to a lesser extent, in normally aging individuals as well ŽMullan and Crawford, 1993; Selkoe, 1994.. These A b-rich aggregates Žalso known as senile plaques. are characteristically infiltrated and surrounded by activated microglia and astrocytes and are usually associated with local cortical atrophy and extensive neuronal loss ŽSelkoe, 1994.. Whereas the
) Corresponding author. Tel.: q39-45-8098130; fax: q39-45-8098127; e-mail: [email protected] 1 Current affiliation: Istituto di Patologia Generale, Universita` di Catania, Italy. 2 Acknowledgement of the equal contribution of the first two authors.
pathogenesis of AD remains poorly understood, the above considerations indicate a potentially important role for A b peptide in the neurodegenerative process. In keeping with this view, both full-length A b and a fragment spanning amino acids 25–35 of the molecule ŽA b Ž25–35.. have been reported to exert a direct cytotoxic effect on neurons in primary hippocampal cultures ŽYankner et al., 1990; Pike et al., 1991, 1993. and to induce neuronal degeneration in adult rat brain in vivo ŽKowall et al., 1991.. In addition to these direct effects, A b can also indirectly promote neuronal damage by virtue of its ability to activate microglia. In this respect, recent studies from our laboratory have demonstrated that co-stimulation of murine microglial cells or human monocytes with A b and IFNg results in the generation of reactive nitrogen intermediates, as well as in the production of pro-inflammatory cytokines such as TNFa , IL-8 and monocyte chemotactic protein-1 ŽMCP-1.rJE ŽMeda et al., 1995, 1996.. Moreover, co-culture experiments established that activation of microglia with A b and IFNg led to neuronal cell injury ŽMeda et al., 1995..
0165-5728r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 5 7 2 8 Ž 9 7 . 0 0 0 5 4 - 4
52
C. Bonaiuto et al.r Journal of Neuroimmunology 77 (1997) 51–56
Despite an increasing body of evidence in support of a role for A b and microglia in neurodegenerative processes, the molecular mechanisms underlying the action of A b towards microglia remain to be elucidated. In this regard, it is noteworthy that the gene encoding the key enzyme involved in nitric oxide generation, namely inducible nitric oxide synthase ŽiNOS., as well as those encoding TNFa and MCP-1rJE, all feature k B or k B-like enhancer sequenceŽs. in their promoter region, which are important for inducible gene expression ŽCollart et al., 1990; Shakhov et al., 1990; Lowenstein et al., 1993; Ueda et al., 1994.. Increased transcriptional activity results from the binding to these enhancer sequences of nuclear factor ŽNF.-k B, an ubiquitous transcription factor normally sequestered in the cytoplasm, but which can be mobilized to the nucleus following appropriate cell stimulation Žfor a review, see Bauerle and Henkel, 1994.. In the present study, we ¨ investigated whether A b peptides have the ability to activate the NF-k B pathway in murine microglia. We now report that A b has the ability to induce NF-k B activation and that co-stimulation of microglial cells or monocytes with IFNg results in an enhancement of this response. This therefore indicates that mobilization of the NF-k B pathway may underlie the action of A b towards cytokine production and inducible nitric oxide generation in microglial cells ŽMeda et al., 1995; Goodwin et al., 1995..
2. Materials and methods 2.1. Reagents and antisera The A b Ž25–35. peptide ŽGSNKGAIIGLM., as well as the full-length A b Ž1–42. peptide, were kindly provided by Dr. L. Otvos ŽWistar Institute, Philadelphia, PA. ŽOtvos et al., 1993.; the peptides were dissolved in sterile PBS, aliquoted and kept at y208C. Recombinant murine IFNg was obtained from Dr. G. Garotta ŽHoffmann-LaRoche, Basel.. Rabbit antisera to human c-Rel Ža265 and a1136, raised against the C-terminal region and against an internal sequence downstream from the nuclear localization signal, respectively., p65rRelA Ža1207 and a1226, against the N-terminal and C-terminal regions, respectively. and p50rNFk B1 Ža1141, against the N-terminal region., were a generous gift from Dr. N.R. Rice ŽNCI-Frederick Cancer Research and Development Center, Frederick, MD.. The specificity of these antisera has already been extensively characterized ŽRice et al., 1992; Rice and Ernst, 1993.. A commercial antibody directed against the nuclear localization sequence of p50 Ža114X. was also purchased from Santa Cruz Biotechnology ŽSanta Cruz, CA.. An oligonucleotide containing tandemly repeated NF-k B sites identical to those of the HIV promoter Ž5X-GATCAGGGACTTTCCGCTGGGGACTTTCC-3X ; underlined is the NF-k B binding sequence. was kindly provided by Dr. G. Trinchieri ŽWistar Institute, Philadelphia, PA.. Poly ŽdI-dC., T4
polynucleotide kinase and Sephadex G25 spin columns were purchased from Pharmacia ŽUppsala, Sweden.. Horseradish peroxidase-linked donkey anti-rabbit antibody, the enhanced chemoluminescence ŽECL. detection kit and w g-32 Px-ATP were purchased from Amersham ŽLittle Chalfont, England.. RPMI 1640 was from GIBCOrBRL ŽGaithersburg, MD. and low-endotoxin FCS Ž- 6 pgrml. from Hyclone ŽLogan, Utah.. Polystyrene flasks and plates for cell culture were from NUNC ŽRotskilde, Denmark.. Aprotinin, acetylated BSA, leupeptin, lipopolysaccharide ŽLPS, from E. coli strain 026B6., Nonidet P40 ŽNP40., pepstatin A and phenylmethanesulphonyl fluoride ŽPMSF. were all from the Sigma Chemical Co. ŽSt. Louis, MO.. All other reagents were molecular biology grade and all buffers and solutions were prepared using pyrogen-free clinical grade water. 2.2. Cell culture The N9 murine microglial cell line was generously provided by Dr. P. Ricciardi-Castagnoli ŽCentro di Citofarmacologia, C.N.R., Milan. ŽRighi et al., 1989.. Microglial cells were cultured in RPMI 1640 supplemented with 10% heat-inactivated FCS, 2 mM L-glutamine, 50 Urml penicillin and 50 m grml streptomycin, at 378C under a 5% CO 2 atmosphere, at a final concentration of 10 6 cellsrml. Rat primary microglia were kindly provided by Dr. U. Armato ŽDepartment of Human Anatomy, University of Verona. and were isolated from Wistar rats and cultured as described previously for murine microglial cells ŽChao et al., 1992.. The rat microglia used in this study were ) 90% pure, as determined by immunocytochemistry. Human peripheral blood monocytes were purified and cultured as described previously ŽMeda et al., 1996.. Cultured cells were exposed to various concentrations of A b Ž25– 35., 50 m grml of ‘aged’ A b Ž1–42. ŽMeda et al., 1995; Pike et al., 1991., or 100 ngrml LPS Žor their diluent, RPMI 1640., for the indicated times. 2.3. Nuclear extract preparation and electrophoretic mobility shift assays After the desired incubation time with the stimuli, microglial cells were washed twice with ice-cold PBS and then harvested using a rubber policeman in 1 ml of lysis buffer Ž10 mM Hepes pH 7.9, 10 mM KCl, 0.2 mM EDTA, 0.2 mM EGTA, 1 mM DTT. supplemented with protease inhibitors Ž0.5 mM PMSF and 10 m grml each of aprotinin, leupeptin and pepstatin A.. After a 15 min incubation on ice, NP40 was added to a final concentration of 0.1% Žvrv. and the samples were vigorously vortexed, prior to centrifugation Ž12,000 g, 1 min, 48C.. Nuclear pellets were resuspended in 30 m l of ice-cold nuclear extraction buffer Ž20 mM Hepes pH 7.9, 400 mM NaCl, 0.2 mM EDTA, 0.2 mM EGTA, 1 mM DTT. containing the antiprotease cocktail and kept under vigorous agitation
C. Bonaiuto et al.r Journal of Neuroimmunology 77 (1997) 51–56
for 15 min at 48C. Samples thus treated were centrifuged Ž12,000 g, 10 min, 48C. and nuclear extracts were aliquoted and stored at y808C. Small aliquots of the extracts were routinely processed for protein content determination. Electrophoretic mobility shift assays ŽEMSA. were performed essentially as described earlier ŽMcDonald et al., 1995.; samples Žcontaining 5 m g of nuclear extract. were electrophoresed on 5% polyacrylamide gels Ž2 h, 200 V, 48C. in 0.5 = TBE and dried gels were exposed to Kodak X-O-Mat film with intensifying screens for periods varying between 6 and 18 h. 2.4. Immunoblots Nuclear extracts were dissolved 2-fold by the addition of an equal volume of sample buffer 2 = Ž50 mM TrisBase pH 6.80, 4% sodium dodecylsulfate Žwrv., 10% 2-mercaptoethanol Žvrv., 20% glycerol Žvrv.., heated for 5 min at 958C and used immediately or stored at y208C. Samples were electrophoresed on 15% Kornberg gels ŽThomas and Kornberg, 1975.. Proteins were then transferred onto a nitrocellulose membrane at 20 V constant for 60 min in a Transblot semidry transfer cell ŽBioRad.. Transfer efficiency was visualized by reversible Ponceau Red staining. The membranes were then processed essentially as described earlier ŽPouliot et al., 1994.. Primary antibodies were used at the following dilutions: anti-RelA Ža1207. or anti-p50 Ža1141., 1:2000; anti c-Rel Ža265., 1:20,000. 3. Results We first examined the ability of A b and of other stimuli to induce NF-k B DNA-binding activity in the
53
murine microglial N9 cell line. For this purpose, cells were cultured in the presence or absence of A b Ž25–35., IFNg , LPS, or combinations thereof, for varying lengths of time. Nuclear proteins were extracted and the NF-k B DNA-binding activity of the extracts was assessed in EMSA. As shown in Fig. 1A, stimulation of microglia with either 100 Urml IFNg or 20 m grml A b Ž25–35. exerted little effect towards NF-k B DNA-binding activity, up to 3 h Žlanes 1, 5, 9 and 2, 6, 10, respectively.. However, co-stimulation of the cells with A b Ž25–35. and IFNg led to a substantial increase in nuclear NF-k B DNA-binding activity, which consisted of a single prominent complex Žband ‘A’. and required at least 120 min to become evident ŽFig. 1A, lane 11.. Similar results were obtained using a commercial A b Ž25–35. peptide ŽSigma., in which the C-terminus was not amidated Ždata not shown.. By contrast, LPS strongly induced the formation of two distinct NF-k B complexes Žbands ‘A’ and ‘B’.; this effect was rapid, as it was already evident after 40 min of stimulation and was sustained for up to 3 h ŽFig. 1A, lanes 4 and 12.. Similar results were obtained in cells stimulated with 100 Urml TNFa Ždata not shown.. In a separate series of experiments, we examined the effect of higher concentrations of A b Ž25–35. towards NF-k B activation in microglial N9 cells. For purposes of comparison, N9 cells were also stimulated with equivalent concentrations of full-length A b Ž1–42., both in the presence and absence of IFNg . Fig. 1B shows that at a concentration of 50 m grml, A b Ž25–35. was able to directly induce NF-k B DNA-binding activity Žlane 2., whereas A b Ž1–42. exerted but a minor effect. Under these conditions, co-stimulation of N9 cells with IFNg markedly enhanced the effect of both A b peptides towards NF-k B activation ŽFig. 1B, lanes 3 and 5..
Fig. 1. ŽA. Time course of NF-k B DNA-binding activity induction by A b Ž25–35., IFNg and LPS in the N9 microglial cell line. Cells were cultured for the indicated times in the presence or absence of 20 m grml A b Ž25–35., 100 Urml IFNg , 100 ngrml LPS, or combinations thereof. Nuclear extracts were then prepared and analyzed in EMSA, as described in Section 2. The amount of nuclear extract used in the binding reactions corresponded to 5 m g of protein. This experiment is representative of four: , unstimulated microglia; A b , A b Ž25–35.-stimulated cells; IFN, IFNg-stimulated cells. ŽB. Comparative effect of A b Ž25–35. and A b Ž1–42. towards nuclear NF-k B DNA-binding activity in IFNg-treated or diluent-treated N9 microglial cells. Cells were cultured for 3 h with 50 m grml A b Ž25–35. or 50 m grml A b Ž1–42., in the presence or absence of 2 ngrml IFNg , prior to EMSA analysis of the resulting nuclear extracts. ctrl, unstimulated microglia; 25–35, A b Ž25–35.-stimulated cells; 1–42, A b Ž1–42.-stimulated cells.
54
C. Bonaiuto et al.r Journal of Neuroimmunology 77 (1997) 51–56
Fig. 2. Characterization of the NF-k B DNA-binding activities present in nuclear extracts of activated N9 microglial cells. ŽA. Nuclear extracts Ž5 m g. from microglial cells stimulated for 3 h with 20 m grml A b Ž25–35. and 2 ngrml IFNg were incubated in the absence Ž‘y’. or in the presence of a 25-fold molar excess of unlabeled oligonucleotide probe Ž‘cold’., prior to the addition of the labeled NF-k B probe to the binding mixtures and subsequent EMSA analysis. ŽB. The same nuclear extracts were incubated in the absence Ž‘y’. or in the presence of specific antisera raised against c-Rel Ža1136., p65rRelA Ža1226., or p50 Ža114X., prior to the addition of the labeled NF-k B probe and subsequent EMSA analysis. As a control, nuclear extracts were also incubated in the presence of pre-immune rabbit serum Ž‘NRS’.. The experiments depicted in this figure are representative of three ŽA. and two ŽB..
We next sought to determine the specificity of the inducible NF-k B complexes detected in microglial cell nuclear extracts. As shown in Fig. 2A, these inducible NF-k B complexes showed specificity in their interaction with our NF-k B oligonucleotide probe, as they were no longer detectable when a 25-fold molar excess of unlabeled probe was included in the binding mixture Žlane 2.; similar results were obtained using nuclear extracts from LPS-treated microglia Ždata not shown.. We also investigated the subunit composition of the inducible NF-k B complexes present in nuclear extracts of activated N9 microglia. To this end, supershift experiments were performed, using specific antisera raised against individual NF-k BrRel proteins. As shown in Fig. 2B, antisera to either RelA or p50 efficiently supershifted the major inducible complex present in extracts from IFNg- plus A b Ž25–35.-treated microglia Žlanes 2 and 3.. In LPStreated cells, the NF-k B DNA-binding activity co-migrating with the aforementioned complex Žband ‘A’. was similarly affected by anti-RelA and anti-p50 antisera Ždata not shown.. By contrast, an antiserum to c-Rel consistently failed to affect the detection or mobility of this NF-k B DNA-binding activity ŽFig. 2B, lane 4.. Our inability to detect c-Rel-containing complexes in supershift experiments prompted us to examine which NF-k BrRel proteins are present in nuclear extracts from activated microglia. Fig. 3 shows that nuclear fractions from unstimulated microglia contain little, if any, immunoreactive p50 or RelA Žlane 1.. Following a 3 h exposure to either IFNg or
A b Ž25–35. as single stimuli, little or no detectable changes in the nuclear levels of p50 or RelA were observed ŽFig. 3, lanes 2 and 3.. However, co-stimulation of microglia with IFNg and A b Ž25–35. consistently resulted in the mobilization of both proteins to the nuclear compartment ŽFig. 3, lane 4.. By comparison, stimulation of microglia for 3 h with LPS led to a stronger nuclear accumulation of both p50 and RelA ŽFig. 3, lane 5.. Under all conditions tested, nuclear c-Rel was undetectable; this could not be attributed to a failure of our c-Rel antiserum to recognize the protein, since immunoreactive c-Rel was readily detected in cytoplasmic extracts Ždata not shown.. In a final series of experiments, we investigated whether the data obtained in the N9 microglial cell line could be reproduced using primary microglial cells. For this purpose, primary rat microglia were isolated and cultured for 3 h in the presence or absence of either 50 m grml A b Ž25–35., 100 Urml IFNg , or both. Fig. 4A shows that by itself, A b Ž25–35. has the ability to substantially induce NF-k B activation Žlane 3. and that co-stimulation of the cells with IFNg resulted in a further Žalbeit modest. enhancement of this response Žlane 4.. Similarly, we investigated whether A b-elicited NF-k B activation might also be observed in cells of human origin. For this purpose, we used human peripheral blood monocytes, since we have previously shown that they respond to A b Ž25–35. in a similar manner to microglia in many respects ŽMeda et al., 1996.. As shown in Fig. 4B, a 3 h culture of human monocytes in the presence of 40 m grml A b Ž25–35. resulted in the induction of a prominent NF-k B DNA-binding activity Žlane 2. and co-stimulation of the cells with IFNg somewhat increased the extent of this response Žlane 4.. Thus, the ability of A b peptides to induce NF-k B activation is not singular feature of cell lines Žsuch as N9
Fig. 3. Nuclear mobilization of NF-k BrRel proteins in response to A b Ž25–35., IFNg and LPS in the N9 microglial cell line. Cells were cultured for 3 h in the presence or absence of 20 m grml A b Ž25–35., 2 ngrml IFNg , 100 ngrml LPS, or combinations thereof. Nuclear extracts were then prepared and analyzed by immunoblotting using antisera to p65rRelA Ža1207. or p50 Ža1141., as described in Section 2. This experiment is representative of at least two. , unstimulated microglia; A b , A b Ž25–35.-stimulated cells; IFN, IFNg-stimulated cells; LPS, LPS-stimulated cells.
C. Bonaiuto et al.r Journal of Neuroimmunology 77 (1997) 51–56
Fig. 4. Activation of NF-k B by A b Ž25–35. and IFNg in rat primary microglia and primary human monocytes. ŽA. Rat primary microglia were cultured for 3 h in the presence or absence of 50 m grml A b Ž25–35. andror 2 ngrml IFNg . Nuclear extracts were then prepared and analyzed in EMSA, as described in Section 2. The amount of nuclear extract used in the binding reactions corresponded to 5 m g of protein. ŽB. Human monocytes were cultured for 3 h in the presence or absence of 40 m grml A b Ž25–35. andror 100 Urml IFNg , prior to EMSA analysis of the resulting nuclear extracts. , unstimulated cells; A b , A b Ž25– 35.-stimulated cells; IFN, IFNg-stimulated cells.
murine microglia., since it was also found to occur in primary rat microglia or primary human monocytes.
4. Discussion In recent years, it has become increasingly evident that both A b and microglial cells play a fundamental role in the pathogenesis of AD. In an effort to elucidate the biological effects of A b towards microglia, we have previously shown that a marked increase in the expression of the transcripts encoding TNFa and MCP-1rJE can be induced in murine N9 microglial cells or in human monocytes by A b Ž25–35. andror IFNg , resulting in the release of the corresponding proteins ŽMeda et al., 1995, 1996.. In view of the fact that NF-k B activation is crucially involved in the inducible expression of the genes encoding TNFa and MCP-1rJE ŽCollart et al., 1990; Shakhov et al., 1990; Ueda et al., 1994., we investigated whether the reported effects of A b Ž25–35. towards microglial gene expression might be paralleled by the activation of NF-k B. We now report that A b Ž25–35. has the ability to activate the NF-k B pathway in murine microglial cells, as well as in primary rat microglia and human monocytes. Although this property of A b Ž25–35. was particularly evident when it was used at higher concentrations, suboptimal concentrations were nevertheless found to synergize with IFNg in inducing NF-k B activation. Immunoblot experiments
55
showed that amongst NF-k BrRel proteins, RelA and p50 are mobilized to the nucleus following microglial cell stimulation with A b Ž25–35. plus IFNg . Accordingly, supershift experiments revealed that the inducible NF-k B DNA-binding activity observed under these conditions consisted of a major complex containing both p50 and RelA. Throughout our study, we also used the well characterized NF-k B activators, LPS and TNFa , for purposes of comparison. As expected, both stimuli were strong inducers of nuclear NF-k B DNA-binding activity, which consisted of at least two specific complexes, one of which being identified as p50rRelA heterodimers. Stimulation of the NF-k B pathway by LPS or TNFa Ži.e. NF-k B activation and the concurrent nuclear mobilization of RelA and p50. was also found to be more rapid and to occur on a greater scale, than that observed in response to A b Ž25–35. even when microglial cells were contemporaneously exposed to A b peptides and IFNg . Taken together, the qualitative and quantitative differences observed between the effects of LPS and those of A b peptides are likely to reflect the existence of distinct signal transduction pathways leading to NF-k B activation. On the basis of the present findings, it is tempting to speculate that NF-k B activation by A b peptides and IFNg might represent one of the molecular mechanisms underlying the inducible expression of the genes encoding TNFa , MCP-1rJE in murine microglia and human monocytes. Similarly, the APP promoter is known to contain two identical k B motifs ŽGrilli et al., 1995. and A b Ž25–35. was recently shown to elicit the transcriptional activation of a k B-driven reporter gene construct in a rat neuronal cell line ŽBehl et al., 1994.. The latter observations therefore raise the possibility, that conditions leading to NF-k B activation in murine microglia might also result in an enhanced expression of the APP gene and of the corresponding protein. In this respect, it is noteworthy that microglia not only constitutively release A b , but can also be induced to secrete substantial quantities of the peptide following stimulation with either LPS or A b Ž25–35. itself ŽBitting et al., 1996.. In a broader context, microglia have been proposed to represent an important source of A b in AD and in view of the above considerations, the combined presence of A b and microglia in senile plaques might contribute to maintain these cells in an activated state. In this context, the ability of A b Ž25–35. to induce the production of TNFa in microglia could lead to a persistent state of inflammation at sites of b-amyloid deposition, and it can be envisaged that the production of MCP-1rJE by activated microglia would further amplify this process through the recruitment of additional microglia to senile plaques. On a final note, we observed that in both murine and rat microglia, A b-elicited NF-k B activation could be markedly enhanced in the presence of IFNg , a cytokine that is primarily derived from activated T-cells and this was especially evident when microglia were exposed to suboptimal concentrations of A b peptides. Although T-
56
C. Bonaiuto et al.r Journal of Neuroimmunology 77 (1997) 51–56
lymphocytes are rarely observed in the central nervous system, they nevertheless have been reported to be present in the vicinity of senile plaques in AD patients, along with microglia ŽRogers et al., 1988; McGeer et al., 1991.. Thus, it is conceivable that microglial cell may activate NF-k B in response to relatively low quantities of A b peptides, through microglialrT-cell interactions. Collectively, the current findings suggest that therapeutical strategies aimed to control the development of AD could include the administration of drugs that hinder NF-k B activation.
Acknowledgements This work was supported by grants to MAC from the Consiglio Nazionale delle Ricerche Žprogetto strategico citochine 95.02809.ST74., from the M.U.R.S.T. Ž40% and 60%.. PPMcD is a post-doctoral Fellow of the Medical Research Council of Canada. We are especially grateful to Dr. Nancy Rice for having provided many antisera; we also wish to thank Dr. Ubaldo Armato for providing rat primary microglia and to Anette Bald, Ercolina Bianchini and Ilaria dal Pra for excellent technical assistance.
References Bauerle, P.A., Henkel, T., 1994. Function and activation of NF-k B in the ¨ immune system. Ann. Rev. Immunol. 12, 141–179. Behl, C., Davis, J.B., Lesley, R., Schubert, D., 1994. Hydrogen peroxide mediates amyloid-b protein toxicity. Cell 77, 817–827. Bitting, L., Naidu, A., Cordell, B., Murphy, G.M., 1996. b-Amyloid peptide secretion by a microglial cell line is induced by b-amyloid Ž25–35. and lipopolysaccharide. J. Biol. Chem. 271, 16084–16089. Chao, C.C., Hu, S., Molitor, T.W., Shaskan, E.G., Peterson, P.K., 1992. Activated microglia mediate neuronal cell injury via a nitric oxide mechanism. J. Immunol. 149, 2736–2741. Collart, M.A., Baeuerle, P., Vassalli, P., 1990. Regulation of tumor necrosis factor-a transcription in macrophages: Involvement of four k B-like motifs and of constitutive and inducible forms of NF-k B. Mol. Cell. Biol. 10, 1498–1506. Goodwin, J.L., Uemura, E., Cunnick, J.E., 1995. Microglial release of nitric oxide by the synergistic action of b-amyloid and IFNg . Brain Res. 692, 207–214. Grilli, M., Ribola, M., Alberici, A., Valerio, A., Memo, M., Spano, P., 1995. Identification and characterization of a k BrRel binding site in the regulatory region of the amyloid precursor protein gene. J. Biol. Chem. 270, 26774–26777. Kowall, N.W., Flint Beal, M., Busciglio, J., Duffy, L.K., Yankner, B.A., 1991. An in vivo model for the neurodegenerative effects of b amyloid and protection by substance P. Proc. Natl. Acad. Sci. USA 88, 7247–7251. Lowenstein, C.J., Alley, E.W., Raval, P., Snowman, A.M., Snyder, S.H.,
Russell, S.W., Murphy, W.J., 1993. Macrophage nitric oxide synthase gene: Two upstream regions mediate induction by interferon-g and lipopolysaccharide. Proc. Natl. Acad. Sci. USA 90, 9730–9734. McDonald, P.P., Cassatella, M.A., Bald, A., Maggi, E., Romagnani, S., Gruss, H.J., Pizzolo, G., 1995. CD30 ligation induces nuclear factork B activation in human T-cell lines. Eur. J. Immunol. 25, 2870–2876. McGeer, P., McGeer, E., Kawamata, H., Yamada, T., Akiyama, H., 1991. Reactions of the immune system in chronic degenerative neurological diseases. Can. J. Neurol. Sci. 18, 376–379. Meda, L., Cassatella, M.A., Szendrei, G.I., Otvos, L., Baron, P., Villalba, M., Ferrari, D., Rossi, F., 1995. Activation of microglial cells by b-amyloid protein and interferon-g . Nature 374, 647–650. Meda, L., Bernasconi, S., Bonaiuto, C., Sozzani, S., Zhou, D., Otvos, L., Mantovani, A., Rossi, F., Cassatella, M.A., 1996. b-Amyloid Ž25–35. peptide and IFN-g synergistically induce the production of the chemotactic cytokine, MCP-1rJE, in monocytes and microglial cells. J. Immunol. 157, 1213–1218. Mullan, M., Crawford, F., 1993. Genetic and molecular advances in Alzheimer’s disease. Trends Neurosci. 16, 398–403. Otvos, L., Szendrei, G.L., Lee, M.V., Mantsch, H.H., 1993. Human and rodent Alzheimer b-amyloid peptides acquire distinct conformations in membrane-mimicking solvents. Eur. J. Biochem. 211, 249–257. Pike, C.J., Walencewicz, A.J., Glabe, C.G., Cotman, C.W., 1991. In vitro aging of b-amyloid protein causes peptide aggregation and neurotoxicity. Brain Res. 563, 311–314. Pike, C.J., Burdick, D., Walencewicz, A.J., Glabe, C.G., Cotman, C.W., 1993. Neurodegeneration induced by b-amyloid peptides in vitro: The role of peptide assembly state. J. Neurosci. 13, 1676–1687. Pouliot, M., McDonald, P.P., Borgeat, P., McColl, S.R., 1994. Granulocyte-macrophage colony-stimulating factor stimulates the expression of the 5-lipoxygenase-activating protein ŽFLAP. in human neutrophils. J. Exp. Med. 179, 1225–1232. Rice, N.R., MacKichan, M.L., Israel, ` A., 1992. The precursor of NF-k B p50 has I k B-like functions. Cell 71, 243–253. Rice, N.R., Ernst, M.K., 1993. In vivo control of NF-k B activation by I k B-a . EMBO J. 12, 4685–4695. Righi, M., Mori, L., De Libero, G., Sironi, M., Biondi, A., Mantovani, A., Donini, S.D., Ricciardi-Castagnoli, P., 1989. Monokine production by microglial cell lines. Eur. J. Immunol. 19, 1443–1448. Rogers, J., Luber Narod, J., Styren, S.D., Civin, W.H., 1988. Expression of immune system-associated antigens by cells of the human central nervous system: Relationship to the pathology of Alzheimer’s disease. Neurobiol. Aging 9, 339–349. Selkoe, D.J., 1994. Cell biology of the amyloid-b protein precursor and the mechanism of Alzheimer’s disease. Annu. Rev. Cell. Biol. 10, 373–403. Shakhov, A.N., Collart, M.A., Vassalli, P., Nedospasov, S.A., Jongeneel, C.V., 1990. k B-type enhancers are involved in lipopolysaccharidemediated transcriptional activation of the tumor necrosis factor-a gene in primary macrophages. J. Exp. Med. 171, 35–47. Thomas, J.O., Kornberg, R.D., 1975. An octamer of histones in chromatin and free in solution. Proc. Natl. Acad. Sci. USA 72, 2626. Ueda, A., Okuda, K., Ohno, S., Shirai, A., Igarashi, T., Matsunaga, K., Fukushima, J., Kawamoto, S., Ishigatsubo, Y., Okubo, T., 1994. NF-k B and Sp1 regulate transcription of the human monocyte chemoattractant protein-1 gene. J. Immunol. 153, 2052–2063. Yankner, B.A., Duffy, L.K., Kirschner, D.A., 1990. Neurotrophic and neurotoxic effects of amyloid-b protein: Reversal by tachykinin neuropeptides. Science 250, 279–282.